- Email: [email protected]
Electroencephalographic characteristics of status epilepticus after cardiac arrest
Clinical Neurophysiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Electroencephalographic characteristics of status epilepticus after cardiac arrest Sofia Backman a,⇑, Erik Westhall a, Irina Dragancea b, Hans Friberg c, Malin Rundgren c, Susann Ullén d, Tobias Cronberg b a
Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Clinical Neurophysiology, Lund, Sweden Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Neurology, Lund, Sweden Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Intensive and Perioperative Care, Lund, Sweden d Clinical Studies Sweden – Forum South, Skane University Hospital, Lund, Sweden b c
a r t i c l e
i n f o
Article history: Accepted 14 January 2017 Available online xxxx Keywords: EEG Continuous EEG monitoring Cardiac arrest Coma Electrographic status epilepticus Hypoxic-ischemic encephalopathy Therapeutic hypothermia Outcome prediction
h i g h l i g h t s Patterns along the ictal-interictal continuum are commonly encountered after cardiac arrest. Unequivocal and possible electrographic status epilepticus (ESE) patients have similar clinical
features. Since ESE patterns appear as a continuum they should not be used in isolation for prognostication.
a b s t r a c t Objective: To describe the electrophysiological characteristics and pathophysiological significance of electrographic status epilepticus (ESE) after cardiac arrest and specifically compare patients with unequivocal ESE to patients with rhythmic or periodic borderline patterns defined as possible ESE. Methods: Retrospective cohort study of consecutive patients treated with targeted temperature management and monitored with simplified continuous EEG. Patients with ESE were identified and electrographically characterised until 72 h after ESE start using the standardised terminology of the American Clinical Neurophysiology Society. Results: ESE occurred in 41 of 127 patients and 22 fulfilled the criteria for unequivocal ESE, which typically appeared early and transiently. Three of the four survivors had unequivocal ESE, starting after rewarming from a continuous background. There were no differences between the groups of unequivocal ESE and possible ESE regarding outcome, neuron-specific enolase levels or prevalence of reported clinical convulsions. Conclusion: ESE is common after cardiac arrest. The distinction between unequivocal and possible ESE patterns was not reflected by differences in clinical features or survival. Significance: A favourable outcome is seen infrequently in patients with ESE, regardless of using strict or liberal ESE definitions. Ó 2017 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Abbreviations: CA, cardiac arrest; cEEG, continuous EEG monitoring; CPC, Cerebral Performance Category; EEG, electroencephalogram; ESE, electrographic status epilepticus; ICU, intensive care unit; IQR, interquartile range; NSE, neuronspecific enolase; ROSC, return of spontaneous circulation; SSEP, somatosensory evoked potentials; WLST, withdrawal of life-sustaining therapy. ⇑ Corresponding author at: Department of Clinical neurophysiology, Skane University Hospital, 221 85 Lund, Sweden. Fax: +46 46 146528. E-mail address: [email protected] (S. Backman).
Hypoxic-ischaemic brain injury is the major cause of death among resuscitated cardiac arrest (CA) patients treated at an intensive care unit (ICU). Electrographic status epilepticus (ESE) is common during the first days of post-arrest care (Rundgren et al., 2006; Rossetti et al., 2007). ESE is recognised as a predictor of poor neurological outcome in recent guidelines, recommending the use of repeated routine EEG or continuous monitoring to detect and treat electrographic seizures (Cronberg et al., 2013; Nolan et al., 2015). Considering the grave consequences of a statement of poor neuro-
http://dx.doi.org/10.1016/j.clinph.2017.01.002 1388-2457/Ó 2017 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
2
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
logical prognosis, often leading to withdrawal of life-sustaining therapy (WLST), it is striking that no agreement exists on the definition of status epilepticus in this setting. Without clear EEG definitions it is a challenging task to decide which patients have a potential for good long-term outcome and for whom prolonged active antiepileptic treatment may be beneficial. The reported prevalence of ESE varies considerably between studies (Rundgren et al., 2010; Legriel et al., 2013; Ruijter et al., 2015), partly due to the usage of diverse definitions, especially regarding criteria for discharge frequency. Several EEG criteria have been proposed (Young et al., 1996; Chong and Hirsch, 2005; Kaplan, 2007; Beniczky et al., 2013; Hirsch et al., 2013) and there is still no international consensus. Gradual changes in electrographic seizure patterns during status epilepticus were described in experimental models and in humans (Treiman et al., 1990; Pender and Losey, 2012). Continuous EEG monitoring (cEEG) may be used to follow these fluctuations along a continuum of rhythmic or periodic patterns ranging from interictal to clearly ictal. The American Clinical Neurophysiology Society (ACNS) has published a standardised terminology for critical care EEG patterns (Hirsch et al., 2013). This terminology includes definitions of periodic and rhythmic patterns that may represent seizure activity, and strict criteria for unequivocal electrographic seizure activity. We have previously described the clinical features, prognostic implications and prevalence of ESE in a well-characterised cohort of CA patients monitored with simplified cEEG (Dragancea et al., 2015). In the present study we used the ACNS terminology to further describe ESE, focusing on the electrophysiological characteristics and development over time. Our main objective was to investigate whether the patients with strictly defined unequivocal ESE differ from patients with possible ESE patterns along the ictalinterictal continuum. 2. Methods 2.1. Patient population and clinical characteristics Retrospective cohort study including consecutive adult CA patients treated with targeted temperature management at the general ICU at Skane University Hospital in Lund between January 2008 and March 2013. Patients were excluded if they had contraindications to temperature management, regained consciousness before start of temperature management, lacked cEEG or follow-up at six months. Patients were intubated, sedated with propofol or midazolam and treated with targeted temperature management at 33 °C or 36 °C for 24 h. Rewarming was completed approximately at 36 h after CA. Sedation was stopped as soon as possible after rewarming. According to local routine, clinical and electrographic seizures were treated by the attending physician with combinations of sedatives and antiepileptic drugs. ESE was actively treated at least until the time-point of prognostication, but without a systematic treatment protocol. The ICU staff noted clinical convulsions including myoclonus on a chart. The study was approved by the Regional Ethical Review Board at Lund University (411/2004, 233/2008, 284/2013). Informed written consent was obtained from next-of-kin and retrospectively from patients who survived. The clinical data and patient characteristics of this cohort were previously published (Dragancea et al., 2015).
two bipolar channels according to the 10–20 system (F3-P3 and F4-P4 or C3-P3 and C4-P4). Filter settings of 1–70 Hz were used. All EEG characterisation was performed by reviewing the original EEG-signal. Amplitude-integrated EEG trend curves were available during analysis, but were not included in the EEG characterisation. All clinical cEEG reports were screened for the presence of epileptiform activity. The cEEG recordings were then retrospectively reviewed by a senior consultant in clinical neurophysiology (SB), blinded to outcome and other clinical parameters. Patients fulfilling the criteria for possible or unequivocal ESE were identified and their cEEG-data was further characterised, from the start of ESE and during the following 72 h. The criteria for unequivocal ESE were based on the ACNS terminology (Hirsch et al., 2013) and considered fulfilled if one of the following patterns occurred: Bilateral spike/sharp-and-waves at a rate of P3 Hz and constituting at least 50% of a 30 min period (Fig. 1a). Repeating sequences of at least 10 s with discharges of any type clearly evolving in frequency, reaching >4 Hz and constituting at least 50% of a 30 min period (Fig. 1b). If the criteria for unequivocal ESE were not fulfilled and rhythmic spike/sharp-and-waves or periodic discharges at a rate of P1 Hz were present for at least 30 min, the pattern was defined as possible ESE (Fig. 1c). The best background pattern was identified during four hours preceding ESE start. The amount of interictal epileptiform discharges during a 30 min period at 12 h and 6 h before ESE start was assessed. Presence of highly epileptiform bursts were noted and defined as bursts with multiple epileptiform discharges with a frequency of P1 Hz in >50% of the bursts (Hirsch et al., 2013). Three consecutive 24-h periods after ESE start were analysed for unequivocal and possible ESE periods. The discharge frequency was assessed and categorised within 0.5 Hz intervals. The dominating and the best background patterns were evaluated for each 24-h period and also for the ESE free periods. Background was assessed for continuity and reported as suppressed (<10 lV peakpeak amplitude), burst-suppression (50–99% suppression), discontinuous (10–49% suppression), nearly continuous (<10% suppression) or continuous according to the ACNS terminology. Only patterns persistent for a minimum of 30 min were reported. According to local practice a routine EEG, recorded with 22 electrodes (10–20 system), was performed before neurological prognostication. The first available routine EEG after ESE start was reviewed retrospectively for background pattern, reactivity and further characterisation of discharges. Reactivity was tested according to a standardised protocol including at least two sound stimulations and at least two pain stimulations (central and peripheral). Bilateral median nerve somatosensory evoked potentials (SSEP) were performed on clinical indication in patients still comatose the day preceding prognostication, typically corresponding to 48–72 h after rewarming. 2.3. Laboratory characteristics Serum levels of neuron-specific enolase (NSE) were analysed at 48 h after CA using NSE Cobas e601 (Roche Diagnostics, Mannheim, Germany). 2.4. Prognostication and criteria for WLST
2.2. Data acquisition and electrophysiological characteristics Patients were monitored with Nicolet One monitors (Viasys Health care, WI, USA) with a simplified cEEG-montage displaying
A neurologist performed prognostication 72 h after rewarming or later according to national recommendations (Cronberg et al., 2013). WLST was considered for comatose patients with Glasgow
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
3
Fig. 1. Electrographic status epilepticus patterns on simplified 2-channel cEEG.
Coma Scale motor score 1–2 in combination with bilateral lack of SSEP N20 potentials or a treatment refractory status epilepticus. If these criteria were not fulfilled the recommendation was to prolong intensive care with daily re-evaluation. 2.5. Follow up A neurologist or an occupational therapist assessed the outcome in a face-to-face meeting six months after CA using the Cerebral Performance Category (CPC) scale. 2.6. Statistical analysis Data was described with median and interquartile range (IQR) or percentages. For comparing differences between groups we used the Fisher’s exact test for categorical data and the Mann–Whitney U-test for continuous data. P < 0.05 was considered statistically significant. Due to the exploratory nature of the study, no corrections for multiple comparisons were made. SPSS version 23.0 was used for statistical analysis. 3. Results 3.1. Patient inclusion The initial cohort consisted of 170 consecutive CA-patients. Of 43 excluded patients, 18 had contraindications to temperature management, 11 regained consciousness before temperature management, 13 lacked cEEG and 1 patient was lost during follow-up (Dragancea et al., 2015). Of the remaining 127 patients, 41 (32%) fulfilled the EEG-criteria for unequivocal or possible ESE and were included in the study. 22 of these 41 patients fulfilled the criteria for unequivocal ESE.
ICU stay. Of 33 patients who underwent SSEP, seven had absent bilateral SSEP N20-peaks and they all died. Median NSE-level at 48 h was 36 ng/mL. 34 out of 41 patients died after WLST, with a median time from CA to death of 8.0 days (IQR 5.0–11.2). Three additional patients died, two of circulatory failure and one of respiratory arrest. Four patients survived to the 6-months follow-up. 3.3. Temporal development of electrographic status epilepticus The median time for initiation of cEEG monitoring was 8 h after CA and the monitoring continued for a median duration of 117 h (Table 1, Fig. 2). No patient had ongoing ESE when the monitoring started and the start of ESE occurred at a median time of 39 h postarrest (Table 1). For 29 of the 41 patients cEEG data was obtained for at least 72 h after ESE start. The reasons for premature cEEG termination were awakening (n = 1), transfer to another department (n = 1), circulatory failure (n = 1), unknown (n = 1) or WLSTdecisions (n = 8). 22 patients fulfilled the strict criteria of unequivocal ESE on at least one time-point and the remaining 19 patients fulfilled possible ESE criteria. The median start time of ESE was 8 h earlier in the unequivocal group compared to the possible group (p = 0.038) (Table 1). The majority of patients with unequivocal ESE (86%) fulfilled these criteria during the first 24 h after start of ESE. Unequivocal ESE was less frequent on day 2 and only three patients had periods with unequivocal ESE during day 3. Most patients in the unequivocal group had possible ESE pattern on day 2 (90%) and day 3 (88%). ESE often ceased 30 min during the review period, also during the first day of ESE (75% of the patients day 1, 71% day 2 and 86% day 3). Prevalence and evolution over time of ESE is further described in Fig. 2 and in Supplementary Table S1, Fig. S1. 3.4. EEG background and discharge characteristics
3.2. Patients characteristics and outcome The characteristics of the ESE-patients are presented in Table 1. The majority (89%) had observed clinical convulsions during the
16 (39%) of the 41 ESE patients had recovered a continuous or nearly continuous background prior to ESE start (Table 2). The proportion of patients with continuous or nearly continuous
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
4
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Table 1 Cohort characteristics. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
33 °C Age Male Initial rhythm shockable (VF/VT) Time to ROSC (min) Clinical convulsions/myoclonus Absent bilateral SSEP N20-peaks NSE at 48 h after CA (ng/mL) Time to initiation of cEEG (h) Time from CA to ESE start (h) cEEG monitoring duration (h) WLST
38/41 (93%) 68 (60–72) 28/41 (68%) 26/40 (65%) 27 (20–40) (n = 39) 36/41 (89%) 7/33 (21%) 36 (23–73) (n = 32) 8 (6–11) 39 (22–48) 117 (75–173) 34/41 (83%)
22/22 (100%) 67 (58–70) 15/22 (68%) 13/21 (62%) 25 (19–36) (n = 21) 19/22 (86%) 3/17 (18%) 35 (22–90) (n = 18) 7 (6–10) 36 (16–44) 110 (80–151) 17/22 (77%)
16/19 (84%) 69 (61–74) 13/19 (68%) 13/19 (68%) 32 (25–42) (n = 18) 17/19 (90%) 4/16 (25%) 41 (29–68) (n = 14) 8 (6–12) 44 (32–68) 137 (60–233) 17/19 (90%)
Outcome at 6 months CPC1 (n) CPC2 (n) CPC3 (n) CPC4 (n) CPC5 (n)
1 2 1 0 37
1 1 1 0 19
0 1 0 0 18
Antiepileptic treatment Monotherapy Combinations* None
11/41 (27%) 29/41 (71%) 1/41 (2%)
4/22 (18%) 18/22 (82%) –
7/19 (37%) 11/19 (58%) 1/19 (5%)
P value
1.000 0.688 0.536 0.038 0.419
Data are presented as number of patients and percentages or median values (IQR). Statistical analysis was performed comparing the unequivocal ESE group and the possible ESE group. Numbers of patients included in the analyses are specified. VT/VF = ventricular tachycardia or fibrillation, ROSC = return of spontaneous circulation, SSEP = somatosensory evoked potentials, NSE = neurone specific enolase, CA = cardiac arrest, ESE = electrographic status epilepticus, WLST = withdrawal of life-sustaining therapy, CPC = Cerebral Performance Category scale. * Antiepileptic treatment with two or three combinations of fosphenytoin, levetiracetam, valproate, phenobarbital, clonazepam and carbamazepine.
Fig. 2. Temporal development of electrographic status epilepticus. Patient 1–22 belong to the unequivocal ESE group and patient 23–41 to the possible ESE group. Starting point and type of ESE pattern is indicated. The following 72 h of monitoring are presented as three 24-h periods representing the worst ESE pattern during that 24-h period. ESE with unequivocal seizures was considered worse than unequivocal ESE with at least 3 Hz discharge frequency. The first vertical line (0 h) represents the time point of cardiac arrest. The second vertical line (36 h after cardiac arrest) indicates the approximate time point of normothermia.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
5
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
background remained unchanged after the start of ESE and throughout the reviewed period (Table 3). The amount of interictal epileptiform discharges and highly epileptiform bursts was comparable at 6 and 12 h before ESE start (Table 2). The discharge morphology during ESE often fluctuated over time between spiky and sharp in the same individual patient. The appearance of the discharges also fluctuated between periodic and rhythmic in many patients. All patients had bilateral synchronous distribution of the discharges during ESE except one who had bilateral discharges at ESE start but developed lateralised discharges after 2 days. 3.5. Comparison between patients with unequivocal and possible ESE There was no difference in the proportion of patients with unequivocal and possible ESE having clinical convulsions (Table 1). The median NSE level at 48 h was elevated in both groups, 35 ng/ mL in the unequivocal group and 41 ng/mL in the possible group, indicating similar burden of hypoxic-ischemic injury. However, only a minority of patients had bilaterally absent SSEP N20responses, similar in both groups. Of the four survivors, three belonged to the unequivocal ESE group. We found a significant difference in EEG background continuity before the ESE start between the two groups, with continuous or nearly continuous background found more often in the possible ESE group (p = 0.029) (Table 2). There were no significant differences in background patterns after ESE start, neither for dominating background, best background or background during ESE-free periods (Table 3). The groups did not differ significantly in the amount of interictal epileptiform discharges before ESE (Table 2). There was a tendency towards higher prevalence of highly epileptiform bursts in the unequivocal group at 6 h before ESE start. 3.6. Subgroup of unequivocal ESE with evolving seizures Unequivocal ESE was defined either by repeating evolving seizures or by high frequency (P3 Hz) spike/sharp-and-waves. ESE with evolving seizures was found in 10 patients, mainly during ongoing target temperature management (Fig. 2). There were no survivors in this subgroup. These patients had a median time from CA to ESE start of 16 h (IQR 14-27) which was shorter compared to the rest of the ESE patients (42 h IQR 32-49, p = 0.001), but the prevalence of clinical convulsions (9 of 10 vs 27 of 31, p = 1.00), median NSE levels (55 ng/mL IQR 22–106 vs 36 23–61, p = 0.624) and the fraction of patients having absent N20-peaks on SSEP (2 of 6 vs 5 of 27, p = 0.584) was similar to the rest. Only one of the
10 patients with evolving seizures had recovered a continuous background prior to ESE start (1 of 10 vs 15 of 31, p = 0.059), but half of the patients had a continuous background as the dominating pattern at some point during the 72 h reviewed period after ESE start. 3.7. Routine EEG A full-montage routine EEG was performed in 35 of the 41 patients. 25 of 35 patients had ongoing possible or unequivocal ESE at this time. A generalised distribution of discharges was seen in 24 of the 25 patients, with frontal predominance in 15 of 24, the remainder having bilateral multifocal discharges. Standardised testing of EEG background reactivity was performed in 24 patients. Preserved reactivity to either sound- or pain-stimuli was seen in 8 of 13 patients in the possible ESE group and 4 of 11 in the unequivocal ESE group. 3.8. Survivors The four survivors had start of ESE after rewarming (Fig. 2). They all regained a continuous background before ESE start, without preceding highly epileptiform bursts and only occasional presence of interictal epileptiform discharges (<1/min). Their best background pattern during the following 72 h was continuous or nearly continuous. The survivors had low or moderately elevated NSE-levels at 48 h (18, 20, 25 and 37 ng/mL) compared to non-survivors (median 47, IQR 26-82 ng/mL) and all four had preserved N20-peaks. Two had clinical convulsions. Three had preserved reactivity to stimuli on the routine EEG whereas the fourth survivor was not tested for reactivity. Three patients had a good outcome at the 6-month follow-up, one with mild or no neurological deficits (CPC 1) and two with moderate disability (CPC 2). One survivor had a severe disability (CPC 3). Three survivors fulfilled the criteria for unequivocal ESE and one had possible ESE. 4. Discussion In our cohort of consecutive CA patients, ESE was a common finding affecting one third of the patients. Half of our ESE patients fulfilled our criteria for unequivocal ESE, based on the ACNS classification, but only temporarily and often during the early phase of ESE as they gradually progressed towards patterns with lower discharge frequencies. This group could not be distinguished,
Table 2 EEG characteristics before start of electrographic status epilepticus. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
P value
Highly epileptiform bursts 12 h before ESE
8/34 (24%)
5/15 (33%)
3/19 (16%)
0.417
Epileptiform discharges 12 h before ESE Abundant (P1/10 s) Frequent (1/min <1/10 s) Occasional/rare (<1/min)
12/33 (36%) 2/33 (6%) 19/33 (58%)
6/14 (43%) 1/14 (7%) 7/14 (50%)
6/19 (32%) 1/19 (5%) 12/19 (63%)
0.854
Highly epileptiform bursts 6 h before ESE
12/41 (29%)
9/22 (41%)
3/19 (16%)
0.098
Epileptiform discharges 6 h before ESE Abundant Frequent Occasional/rare
17/40 (43%) 5/40 (12%) 18/40 (45%)
9/22 (41%) 2/22 (9%) 11/22 (50%)
8/18 (44%) 3/18 (17%) 7/18 (39%)
0.748
Best background 0–4 h before ESE Continuous/nearly continuous background
16/41 (39%)
5/22 (23%)
11/19 (58%)
0.029
Occurrence of epileptiform discharges and continuous/nearly continuous background prior to ESE start. The amount of interictal epileptiform discharges was categorised according to the ACNS quantification of non-rhythmic non-periodic discharges. Data are presented as number of patients and percentages. Unequivocal and possible ESE groups are compared statistically. Numbers of patients included in the analyses are specified.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
6
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Table 3 EEG background day 1–3 after start of electrographic status epilepticus. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
P value
Continuous/nearly continuous as dominating background 0–24 h after ESE start 16/36 (44%) 24–48 h after ESE start 13/29 (45%) 48–72 h after ESE start 14/29 (48%)
7/19 (37%) 6/16 (38%) 7/16 (44%)
9/17 (53%) 7/13 (54%) 7/13 (54%)
0.503 0.467 0.715
Continuous/nearly continuous as best background 0–24 h after ESE start 20/37 (54%) 24–48 h after ESE start 19/30 (63%) 48–72 h after ESE start 16/29 (55%)
8/19 (42%) 8/16 (50%) 9/16 (56%)
12/18 (67%) 11/14 (79%) 7/13 (54%)
0.191 0.142 1.000
Continuous/nearly continuous as best background during ESE-free periods 0–24 h after ESE start 17/31 (55%) 24–48 h after ESE start 17/25 (68%) 48–72 h after ESE start 15/25 (60%)
6/14 (43%) 7/12 (58%) 8/13 (62%)
11/17 (65%) 10/13 (77%) 7/12 (58%)
0.289 0.411 1.000
Patients with a continuous/nearly continuous background, with or without ongoing ESE during the three 24-h periods following ESE start. The number of patients with interpretable cEEG background patterns is specified. Missing data is due to shorter monitoring periods, artefacts or persistent high discharge frequencies making the background interpretation uncertain. Unequivocal and possible ESE groups are compared statistically.
regarding clinical features including markers of brain injury, from the patients only showing possible ESE patterns. In clinical practise any pattern along the ictal-interictal continuum may or may not represent seizure activity in the individual patient. Previously used discharge frequencies for defining ESE after CA span from 0.5 Hz to 3 Hz (Hirsch et al., 2013; Amorim et al., 2015; Ruijter et al., 2015; Sivaraju and Gilmore, 2016). We chose a threshold for possible ESE in the lower range (1 Hz) to include most patterns in the continuum. The recent Salzburg Consensus Criteria for diagnosis of nonconvulsive status epilepticus proposed the term possible status epilepticus (Beniczky et al., 2013; Leitinger et al., 2016). If a possible ESE pattern occur in close relation to subtle clinical twitching it is considered diagnostic for status epilepticus. Importantly, the majority (90%) of our patients with possible ESE had clinical convulsions or myoclonus at some point after CA. A limitation with our retrospective data is that we cannot confirm temporal relationship between convulsions and EEG-discharges due to lack of video-EEG registrations. By comparing EEG characteristics of patients fulfilling our liberal definition of possible ESE with those having unequivocal ESE patterns, we aimed to investigate whether such a distinction had any bearing on prognosis. In addition, we used the serum level of NSE and the SSEP N20-responses as surrogate markers of brain injury. A correlation between NSE, EEG and other markers of postanoxic brain injury has previously been demonstrated (Cronberg et al., 2011; Rossetti et al., 2012). NSE levels were moderately or highly elevated in most patients reflecting the underlying brain injury. Interestingly, the majority of our patients in both ESE groups had present SSEP N20-peaks contradicting a very pronounced brain injury, but consistent with the notion that at least partial cortical integrity is necessary for seizure activity to occur. Development of ESE after cardiac arrest is per se a sign of poor prognosis (Rossetti et al., 2007) and the majority of these patients die, as in our cohort. An early start of ESE correlates with worse outcome (Ruijter et al., 2015). Patients who develop a continuous background prior to ESE start have been found to have a somewhat better outcome and lower NSE-levels than patients with a discontinuous or suppressed background (Rundgren et al., 2010; Dragancea et al., 2015). Our findings that the unequivocal ESE group had earlier ESE start, less often preceded by a continuous background, may indicate a more extensive brain damage. We could not support this hypothesis by any differences in other markers of brain injury such as NSE-levels, SSEP N20-potentials, presence of clinical convulsions or by their final outcome. In addition we found no differences in the EEG background continuity between the two electrographic categories once ESE had started,
but we recognise that it is difficult to define the background pattern during ongoing ESE. The subgroup of unequivocal ESE with repeating evolving seizures had similar clinical characteristics compared to the other ESE-patients. Yet, these patients had a significantly earlier start of ESE, a tendency towards a more discontinuous background prior to ESE start and none survived. Further studies with larger cohorts are needed to investigate whether this sub-categorisation has any prognostic information to add and whether it represents a distinct pathophysiological phenomenon. Although our cohort is large in comparison, the number of patients in some subgroups is small (survivors, evolving seizures). Therefore comparisons between these subgroups are limited by the lack of power and differences may have been missed. Importantly, three of our four survivors belonged to the unequivocal ESE group demonstrating that this strict definition of ESE cannot reliably identify patients with poor prognosis. None of the survivors had evolving seizures or highly epileptiform bursts. The survivors had a continuous and reactive background and were classified as unequivocal ESE based on the high discharge frequency. A similar EEG-pattern was recently described among patients surviving despite early postanoxic multifocal myoclonus (Elmer et al., 2016). In our cohort we only had four survivors with ESE, and they were all treated for a prolonged period in the ICU (16–24 days). In post cardiac arrest studies there is always a concern about the self-fulfilling prophecy. We used a multimodal protocol for prognostication at a delayed time-point. Patients with possible and unequivocal ESE were actively treated typically with increased sedation and combinations of antiepileptic agents, at least until prognostication. Whether additional patients in our cohort could have benefitted from even more aggressive and prolonged antiepileptic treatment is outside the scope of this study. We emphasize that the present study was not designed to investigate whether there are differences in the response to antiepileptic treatment among patients with unequivocal seizure patterns compared to those with possible ESE patterns. We also acknowledge that the degree of sedation may have contributed to changes in background patterns in some of our patients. Therefore we also reported the best achieved background pattern. On a group level, a recent routine EEG study showed no significant impact of ongoing sedation, in clinically used doses, on the predictive value of EEG after CA (Westhall et al., 2016). For a standard cEEG the use of at least 16 electrodes is recommended (Herman et al., 2015), but this is resource consuming and challenging to perform outside large centres. We monitored
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
our patients with a simplified EEG montage using a limited number of electrodes and acknowledge that occasional patients with seizures may be missed by this approach. The sensitivity to detect seizure activity with a simplified montage in patients with mixed aetiologies is lower (68–93%) compared to a multi-channel EEG montage (Kolls and Husain, 2007; Young et al., 2009; Karakis et al., 2010; Rubin et al., 2014). However, in adult post cardiac arrest patients the discharges most frequently show generalised distribution (Westhall et al., 2015) and a recent study on postcardiac arrest patients with simplified montage showed a substantially higher yield (92% or 100% depending on montage) (Vanherpe and Schrooten, 2016). Furthermore, the routine EEGs in our cohort often showed generalised discharges with a frontal predominance, which are most likely to be detected using our simplified montage. A reduced number of electrodes with larger inter-electrode distance may lead to slightly higher EEG amplitudes. Compared to other investigators using full-montage cEEG, we may therefore have been more prone to classify background patterns as benign (Tjepkema-Cloostermans et al., 2016). Despite these limitations our findings highlight that ESE patterns should be regarded as a continuum and that the definition of unequivocal ESE patterns based on discharge frequencies appears to be of minor importance for prognostication and outcome. Instead a multimodal prognostic approach is advocated (Callaway et al., 2015) which could also be used as a strategy for identifying those ESE patients with a potential for recovery (Rossetti et al., 2010; Westhall et al., 2013; Dragancea et al., 2015). 5. Conclusion EEG patterns compatible with status epilepticus after CA is a common finding in the ICU and have varying electrographic characteristics. Further distinction into unequivocal status epilepticus, based on the ACNS-classification, is in this study not supported by other features such as biomarkers, clinical characteristics or outcome and may therefore be of minor clinical relevance for prognostication. To support clinical decisions, a multimodal prognostication strategy incorporating other aspects of the EEG, for instance background reactivity and continuity, is advocated. Funding source Supported by the Skane University Hospital foundation, the Swedish National Health System (ALF), the County Council of Skane and the Koch Foundation. Acknowledgements The authors would like to thank Professor Ingmar Rosén (Department of Clinical Neurophysiology, Skane University Hospital, Lund, Sweden) for his valuable guidance. Conflict of interest: None of the authors have potential conflicts of interest to be disclosed. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.clinph.2017.01. 002. References Amorim E, Rittenberger JC, Baldwin ME, Callaway CW, Popescu APost Cardiac Arrest S. Malignant EEG patterns in cardiac arrest patients treated with targeted temperature management who survive to hospital discharge. Resuscitation 2015;90:127–32.
7
Beniczky S, Hirsch LJ, Kaplan PW, Pressler R, Bauer G, Aurlien H, et al. Unified EEG terminology and criteria for nonconvulsive status epilepticus. Epilepsia 2013;54 (suppl 6):28–9. Callaway CW, Soar J, Aibiki M, Bottiger BW, Brooks SC, Deakin CD, et al. Part 4: Advanced Life Support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S84–145. Chong DJ, Hirsch LJ. Which EEG patterns warrant treatment in the critically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005;22:79–91. Cronberg T, Brizzi M, Liedholm LJ, Rosen I, Rubertsson S, Rylander C, et al. Neurological prognostication after cardiac arrest–recommendations from the Swedish Resuscitation Council. Resuscitation 2013;84:867–72. Cronberg T, Rundgren M, Westhall E, Englund E, Siemund R, Rosen I, et al. Neuronspecific enolase correlates with other prognostic markers after cardiac arrest. Neurology 2011;77:623–30. Dragancea I, Backman S, Westhall E, Rundgren M, Friberg H, Cronberg T. Outcome following postanoxic status epilepticus in patients with targeted temperature management after cardiac arrest. Epilepsy Behav 2015;49:173–7. Elmer J, Rittenberger JC, Faro J, Molyneaux BJ, Popescu A, Callaway CW, et al. Clinically distinct electroencephalographic phenotypes of early myoclonus after cardiac arrest. Ann Neurol 2016;80:175–84. Herman ST, Abend NS, Bleck TP, Chapman KE, Drislane FW, Emerson RG, et al. Consensus statement on continuous EEG in critically ill adults and children, part II: personnel, technical specifications, and clinical practice. J Clin Neurophysiol 2015;32:96–108. Hirsch LJ, LaRoche SM, Gaspard N, Gerard E, Svoronos A, Herman ST, et al. American clinical neurophysiology society’s standardized critical care EEG terminology: 2012 version. J Clin Neurophysiol 2013;30:1–27. Kaplan PW. EEG criteria for nonconvulsive status epilepticus. Epilepsia 2007;48 (suppl 8):39–41. Karakis I, Montouris GD, Otis JA, Douglass LM, Jonas R, Velez-Ruiz N, et al. A quick and reliable EEG montage for the detection of seizures in the critical care setting. J Clin Neurophysiol 2010;27:100–5. Kolls BJ, Husain AM. Assessment of hairline EEG as a screening tool for nonconvulsive status epilepticus. Epilepsia 2007;48:959–65. Legriel S, Hilly-Ginoux J, Resche-Rigon M, Merceron S, Pinoteau J, Henry-Lagarrigue M, et al. Prognostic value of electrographic postanoxic status epilepticus in comatose cardiac-arrest survivors in the therapeutic hypothermia era. Resuscitation 2013;84:343–50. Leitinger M, Trinka E, Gardella E, Rohracher A, Kalss G, Qerama E, et al. Diagnostic accuracy of the Salzburg EEG criteria for non-convulsive status epilepticus: a retrospective study. Lancet Neurol 2016;15:1054–62. Nolan JP, Soar J, Cariou A, Cronberg T, Moulaert VR, Deakin CD, et al. European resuscitation council and european society of intensive care medicine guidelines for post-resuscitation care 2015: Section 5 of the European Resuscitation Council Guidelines for Resuscitation 2015. Resuscitation 2015;95:202–22. Pender RA, Losey TE. A rapid course through the five electrographic stages of status epilepticus. Epilepsia 2012;53:e193–5. Rossetti AO, Carrera E, Oddo M. Early EEG correlates of neuronal injury after brain anoxia. Neurology 2012;78:796–802. Rossetti AO, Logroscino G, Liaudet L, Ruffieux C, Ribordy V, Schaller MD, et al. Status epilepticus: an independent outcome predictor after cerebral anoxia. Neurology 2007;69:255–60. Rossetti AO, Oddo M, Logroscino G, Kaplan PW. Prognostication after cardiac arrest and hypothermia: a prospective study. Ann Neurol 2010;67:301–7. Rubin MN, Jeffery OJ, Fugate JE, Britton JW, Cascino GD, Worrell GA, et al. Efficacy of a reduced electroencephalography electrode array for detection of seizures. Neurohospitalist 2014;4:6–8. Ruijter BJ, van Putten MJ, Hofmeijer J. Generalized epileptiform discharges in postanoxic encephalopathy: quantitative characterization in relation to outcome. Epilepsia 2015;56:1845–54. Rundgren M, Rosen I, Friberg H. Amplitude-integrated EEG (aEEG) predicts outcome after cardiac arrest and induced hypothermia. Intensive Care Med 2006;32:836–42. Rundgren M, Westhall E, Cronberg T, Rosen I, Friberg H. Continuous amplitudeintegrated electroencephalogram predicts outcome in hypothermia-treated cardiac arrest patients. Crit Care Med 2010;38:1838–44. Sivaraju A, Gilmore EJ. Understanding and managing the ictal-interictal continuum in neurocritical care. Curr Treat Options Neurol 2016;18:8. Tjepkema-Cloostermans MC, Hofmeijer J, Hom HW, Bosch FH, van Putten MJ. Predicting outcome in postanoxic coma: are ten EEG electrodes enough? J Clin Neurophysiol 2016. http://dx.doi.org/10.1097/WNP.0000000000000337. Treiman DM, Walton NY, Kendrick C. A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus. Epilepsy Res 1990;5:49–60. Vanherpe P, Schrooten M. Minimal EEG montage with high yield for the detection of status epilepticus in the setting of postanoxic brain damage. Acta Neurol Belg 2016. http://dx.doi.org/10.1007/s13760-016-0663-9. Westhall E, Rosen I, Rossetti AO, van Rootselaar AF, Wesenberg Kjaer T, Friberg H, et al. Interrater variability of EEG interpretation in comatose cardiac arrest patients. Clin Neurophysiol 2015;126:2397–404. Westhall E, Rossetti AO, van Rootselaar AF, Wesenberg Kjaer T, Horn J, Ullen S, et al. Standardized EEG interpretation accurately predicts prognosis after cardiac arrest. Neurology 2016;86:1482–90.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
8
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Westhall E, Rundgren M, Lilja G, Friberg H, Cronberg T. Postanoxic status epilepticus can be identified and treatment guided successfully by continuous electroencephalography. Ther Hypothermia Temp Manage 2013;3:84–7. Young GB, Jordan KG, Doig GS. An assessment of nonconvulsive seizures in the intensive care unit using continuous EEG monitoring: an investigation of variables associated with mortality. Neurology 1996;47:83–9.
Young GB, Sharpe MD, Savard M, Al Thenayan E, Norton L, Davies-Schinkel C. Seizure detection with a commercially available bedside EEG monitor and the subhairline montage. Neurocrit Care 2009;11:411–6.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Electroencephalographic characteristics of status epilepticus after cardiac arrest Sofia Backman a,⇑, Erik Westhall a, Irina Dragancea b, Hans Friberg c, Malin Rundgren c, Susann Ullén d, Tobias Cronberg b a
Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Clinical Neurophysiology, Lund, Sweden Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Neurology, Lund, Sweden Lund University, Skane University Hospital, Department of Clinical Sciences Lund, Intensive and Perioperative Care, Lund, Sweden d Clinical Studies Sweden – Forum South, Skane University Hospital, Lund, Sweden b c
a r t i c l e
i n f o
Article history: Accepted 14 January 2017 Available online xxxx Keywords: EEG Continuous EEG monitoring Cardiac arrest Coma Electrographic status epilepticus Hypoxic-ischemic encephalopathy Therapeutic hypothermia Outcome prediction
h i g h l i g h t s Patterns along the ictal-interictal continuum are commonly encountered after cardiac arrest. Unequivocal and possible electrographic status epilepticus (ESE) patients have similar clinical
features. Since ESE patterns appear as a continuum they should not be used in isolation for prognostication.
a b s t r a c t Objective: To describe the electrophysiological characteristics and pathophysiological significance of electrographic status epilepticus (ESE) after cardiac arrest and specifically compare patients with unequivocal ESE to patients with rhythmic or periodic borderline patterns defined as possible ESE. Methods: Retrospective cohort study of consecutive patients treated with targeted temperature management and monitored with simplified continuous EEG. Patients with ESE were identified and electrographically characterised until 72 h after ESE start using the standardised terminology of the American Clinical Neurophysiology Society. Results: ESE occurred in 41 of 127 patients and 22 fulfilled the criteria for unequivocal ESE, which typically appeared early and transiently. Three of the four survivors had unequivocal ESE, starting after rewarming from a continuous background. There were no differences between the groups of unequivocal ESE and possible ESE regarding outcome, neuron-specific enolase levels or prevalence of reported clinical convulsions. Conclusion: ESE is common after cardiac arrest. The distinction between unequivocal and possible ESE patterns was not reflected by differences in clinical features or survival. Significance: A favourable outcome is seen infrequently in patients with ESE, regardless of using strict or liberal ESE definitions. Ó 2017 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Abbreviations: CA, cardiac arrest; cEEG, continuous EEG monitoring; CPC, Cerebral Performance Category; EEG, electroencephalogram; ESE, electrographic status epilepticus; ICU, intensive care unit; IQR, interquartile range; NSE, neuronspecific enolase; ROSC, return of spontaneous circulation; SSEP, somatosensory evoked potentials; WLST, withdrawal of life-sustaining therapy. ⇑ Corresponding author at: Department of Clinical neurophysiology, Skane University Hospital, 221 85 Lund, Sweden. Fax: +46 46 146528. E-mail address: [email protected] (S. Backman).
Hypoxic-ischaemic brain injury is the major cause of death among resuscitated cardiac arrest (CA) patients treated at an intensive care unit (ICU). Electrographic status epilepticus (ESE) is common during the first days of post-arrest care (Rundgren et al., 2006; Rossetti et al., 2007). ESE is recognised as a predictor of poor neurological outcome in recent guidelines, recommending the use of repeated routine EEG or continuous monitoring to detect and treat electrographic seizures (Cronberg et al., 2013; Nolan et al., 2015). Considering the grave consequences of a statement of poor neuro-
http://dx.doi.org/10.1016/j.clinph.2017.01.002 1388-2457/Ó 2017 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
2
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
logical prognosis, often leading to withdrawal of life-sustaining therapy (WLST), it is striking that no agreement exists on the definition of status epilepticus in this setting. Without clear EEG definitions it is a challenging task to decide which patients have a potential for good long-term outcome and for whom prolonged active antiepileptic treatment may be beneficial. The reported prevalence of ESE varies considerably between studies (Rundgren et al., 2010; Legriel et al., 2013; Ruijter et al., 2015), partly due to the usage of diverse definitions, especially regarding criteria for discharge frequency. Several EEG criteria have been proposed (Young et al., 1996; Chong and Hirsch, 2005; Kaplan, 2007; Beniczky et al., 2013; Hirsch et al., 2013) and there is still no international consensus. Gradual changes in electrographic seizure patterns during status epilepticus were described in experimental models and in humans (Treiman et al., 1990; Pender and Losey, 2012). Continuous EEG monitoring (cEEG) may be used to follow these fluctuations along a continuum of rhythmic or periodic patterns ranging from interictal to clearly ictal. The American Clinical Neurophysiology Society (ACNS) has published a standardised terminology for critical care EEG patterns (Hirsch et al., 2013). This terminology includes definitions of periodic and rhythmic patterns that may represent seizure activity, and strict criteria for unequivocal electrographic seizure activity. We have previously described the clinical features, prognostic implications and prevalence of ESE in a well-characterised cohort of CA patients monitored with simplified cEEG (Dragancea et al., 2015). In the present study we used the ACNS terminology to further describe ESE, focusing on the electrophysiological characteristics and development over time. Our main objective was to investigate whether the patients with strictly defined unequivocal ESE differ from patients with possible ESE patterns along the ictalinterictal continuum. 2. Methods 2.1. Patient population and clinical characteristics Retrospective cohort study including consecutive adult CA patients treated with targeted temperature management at the general ICU at Skane University Hospital in Lund between January 2008 and March 2013. Patients were excluded if they had contraindications to temperature management, regained consciousness before start of temperature management, lacked cEEG or follow-up at six months. Patients were intubated, sedated with propofol or midazolam and treated with targeted temperature management at 33 °C or 36 °C for 24 h. Rewarming was completed approximately at 36 h after CA. Sedation was stopped as soon as possible after rewarming. According to local routine, clinical and electrographic seizures were treated by the attending physician with combinations of sedatives and antiepileptic drugs. ESE was actively treated at least until the time-point of prognostication, but without a systematic treatment protocol. The ICU staff noted clinical convulsions including myoclonus on a chart. The study was approved by the Regional Ethical Review Board at Lund University (411/2004, 233/2008, 284/2013). Informed written consent was obtained from next-of-kin and retrospectively from patients who survived. The clinical data and patient characteristics of this cohort were previously published (Dragancea et al., 2015).
two bipolar channels according to the 10–20 system (F3-P3 and F4-P4 or C3-P3 and C4-P4). Filter settings of 1–70 Hz were used. All EEG characterisation was performed by reviewing the original EEG-signal. Amplitude-integrated EEG trend curves were available during analysis, but were not included in the EEG characterisation. All clinical cEEG reports were screened for the presence of epileptiform activity. The cEEG recordings were then retrospectively reviewed by a senior consultant in clinical neurophysiology (SB), blinded to outcome and other clinical parameters. Patients fulfilling the criteria for possible or unequivocal ESE were identified and their cEEG-data was further characterised, from the start of ESE and during the following 72 h. The criteria for unequivocal ESE were based on the ACNS terminology (Hirsch et al., 2013) and considered fulfilled if one of the following patterns occurred: Bilateral spike/sharp-and-waves at a rate of P3 Hz and constituting at least 50% of a 30 min period (Fig. 1a). Repeating sequences of at least 10 s with discharges of any type clearly evolving in frequency, reaching >4 Hz and constituting at least 50% of a 30 min period (Fig. 1b). If the criteria for unequivocal ESE were not fulfilled and rhythmic spike/sharp-and-waves or periodic discharges at a rate of P1 Hz were present for at least 30 min, the pattern was defined as possible ESE (Fig. 1c). The best background pattern was identified during four hours preceding ESE start. The amount of interictal epileptiform discharges during a 30 min period at 12 h and 6 h before ESE start was assessed. Presence of highly epileptiform bursts were noted and defined as bursts with multiple epileptiform discharges with a frequency of P1 Hz in >50% of the bursts (Hirsch et al., 2013). Three consecutive 24-h periods after ESE start were analysed for unequivocal and possible ESE periods. The discharge frequency was assessed and categorised within 0.5 Hz intervals. The dominating and the best background patterns were evaluated for each 24-h period and also for the ESE free periods. Background was assessed for continuity and reported as suppressed (<10 lV peakpeak amplitude), burst-suppression (50–99% suppression), discontinuous (10–49% suppression), nearly continuous (<10% suppression) or continuous according to the ACNS terminology. Only patterns persistent for a minimum of 30 min were reported. According to local practice a routine EEG, recorded with 22 electrodes (10–20 system), was performed before neurological prognostication. The first available routine EEG after ESE start was reviewed retrospectively for background pattern, reactivity and further characterisation of discharges. Reactivity was tested according to a standardised protocol including at least two sound stimulations and at least two pain stimulations (central and peripheral). Bilateral median nerve somatosensory evoked potentials (SSEP) were performed on clinical indication in patients still comatose the day preceding prognostication, typically corresponding to 48–72 h after rewarming. 2.3. Laboratory characteristics Serum levels of neuron-specific enolase (NSE) were analysed at 48 h after CA using NSE Cobas e601 (Roche Diagnostics, Mannheim, Germany). 2.4. Prognostication and criteria for WLST
2.2. Data acquisition and electrophysiological characteristics Patients were monitored with Nicolet One monitors (Viasys Health care, WI, USA) with a simplified cEEG-montage displaying
A neurologist performed prognostication 72 h after rewarming or later according to national recommendations (Cronberg et al., 2013). WLST was considered for comatose patients with Glasgow
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
3
Fig. 1. Electrographic status epilepticus patterns on simplified 2-channel cEEG.
Coma Scale motor score 1–2 in combination with bilateral lack of SSEP N20 potentials or a treatment refractory status epilepticus. If these criteria were not fulfilled the recommendation was to prolong intensive care with daily re-evaluation. 2.5. Follow up A neurologist or an occupational therapist assessed the outcome in a face-to-face meeting six months after CA using the Cerebral Performance Category (CPC) scale. 2.6. Statistical analysis Data was described with median and interquartile range (IQR) or percentages. For comparing differences between groups we used the Fisher’s exact test for categorical data and the Mann–Whitney U-test for continuous data. P < 0.05 was considered statistically significant. Due to the exploratory nature of the study, no corrections for multiple comparisons were made. SPSS version 23.0 was used for statistical analysis. 3. Results 3.1. Patient inclusion The initial cohort consisted of 170 consecutive CA-patients. Of 43 excluded patients, 18 had contraindications to temperature management, 11 regained consciousness before temperature management, 13 lacked cEEG and 1 patient was lost during follow-up (Dragancea et al., 2015). Of the remaining 127 patients, 41 (32%) fulfilled the EEG-criteria for unequivocal or possible ESE and were included in the study. 22 of these 41 patients fulfilled the criteria for unequivocal ESE.
ICU stay. Of 33 patients who underwent SSEP, seven had absent bilateral SSEP N20-peaks and they all died. Median NSE-level at 48 h was 36 ng/mL. 34 out of 41 patients died after WLST, with a median time from CA to death of 8.0 days (IQR 5.0–11.2). Three additional patients died, two of circulatory failure and one of respiratory arrest. Four patients survived to the 6-months follow-up. 3.3. Temporal development of electrographic status epilepticus The median time for initiation of cEEG monitoring was 8 h after CA and the monitoring continued for a median duration of 117 h (Table 1, Fig. 2). No patient had ongoing ESE when the monitoring started and the start of ESE occurred at a median time of 39 h postarrest (Table 1). For 29 of the 41 patients cEEG data was obtained for at least 72 h after ESE start. The reasons for premature cEEG termination were awakening (n = 1), transfer to another department (n = 1), circulatory failure (n = 1), unknown (n = 1) or WLSTdecisions (n = 8). 22 patients fulfilled the strict criteria of unequivocal ESE on at least one time-point and the remaining 19 patients fulfilled possible ESE criteria. The median start time of ESE was 8 h earlier in the unequivocal group compared to the possible group (p = 0.038) (Table 1). The majority of patients with unequivocal ESE (86%) fulfilled these criteria during the first 24 h after start of ESE. Unequivocal ESE was less frequent on day 2 and only three patients had periods with unequivocal ESE during day 3. Most patients in the unequivocal group had possible ESE pattern on day 2 (90%) and day 3 (88%). ESE often ceased 30 min during the review period, also during the first day of ESE (75% of the patients day 1, 71% day 2 and 86% day 3). Prevalence and evolution over time of ESE is further described in Fig. 2 and in Supplementary Table S1, Fig. S1. 3.4. EEG background and discharge characteristics
3.2. Patients characteristics and outcome The characteristics of the ESE-patients are presented in Table 1. The majority (89%) had observed clinical convulsions during the
16 (39%) of the 41 ESE patients had recovered a continuous or nearly continuous background prior to ESE start (Table 2). The proportion of patients with continuous or nearly continuous
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
4
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Table 1 Cohort characteristics. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
33 °C Age Male Initial rhythm shockable (VF/VT) Time to ROSC (min) Clinical convulsions/myoclonus Absent bilateral SSEP N20-peaks NSE at 48 h after CA (ng/mL) Time to initiation of cEEG (h) Time from CA to ESE start (h) cEEG monitoring duration (h) WLST
38/41 (93%) 68 (60–72) 28/41 (68%) 26/40 (65%) 27 (20–40) (n = 39) 36/41 (89%) 7/33 (21%) 36 (23–73) (n = 32) 8 (6–11) 39 (22–48) 117 (75–173) 34/41 (83%)
22/22 (100%) 67 (58–70) 15/22 (68%) 13/21 (62%) 25 (19–36) (n = 21) 19/22 (86%) 3/17 (18%) 35 (22–90) (n = 18) 7 (6–10) 36 (16–44) 110 (80–151) 17/22 (77%)
16/19 (84%) 69 (61–74) 13/19 (68%) 13/19 (68%) 32 (25–42) (n = 18) 17/19 (90%) 4/16 (25%) 41 (29–68) (n = 14) 8 (6–12) 44 (32–68) 137 (60–233) 17/19 (90%)
Outcome at 6 months CPC1 (n) CPC2 (n) CPC3 (n) CPC4 (n) CPC5 (n)
1 2 1 0 37
1 1 1 0 19
0 1 0 0 18
Antiepileptic treatment Monotherapy Combinations* None
11/41 (27%) 29/41 (71%) 1/41 (2%)
4/22 (18%) 18/22 (82%) –
7/19 (37%) 11/19 (58%) 1/19 (5%)
P value
1.000 0.688 0.536 0.038 0.419
Data are presented as number of patients and percentages or median values (IQR). Statistical analysis was performed comparing the unequivocal ESE group and the possible ESE group. Numbers of patients included in the analyses are specified. VT/VF = ventricular tachycardia or fibrillation, ROSC = return of spontaneous circulation, SSEP = somatosensory evoked potentials, NSE = neurone specific enolase, CA = cardiac arrest, ESE = electrographic status epilepticus, WLST = withdrawal of life-sustaining therapy, CPC = Cerebral Performance Category scale. * Antiepileptic treatment with two or three combinations of fosphenytoin, levetiracetam, valproate, phenobarbital, clonazepam and carbamazepine.
Fig. 2. Temporal development of electrographic status epilepticus. Patient 1–22 belong to the unequivocal ESE group and patient 23–41 to the possible ESE group. Starting point and type of ESE pattern is indicated. The following 72 h of monitoring are presented as three 24-h periods representing the worst ESE pattern during that 24-h period. ESE with unequivocal seizures was considered worse than unequivocal ESE with at least 3 Hz discharge frequency. The first vertical line (0 h) represents the time point of cardiac arrest. The second vertical line (36 h after cardiac arrest) indicates the approximate time point of normothermia.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
5
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
background remained unchanged after the start of ESE and throughout the reviewed period (Table 3). The amount of interictal epileptiform discharges and highly epileptiform bursts was comparable at 6 and 12 h before ESE start (Table 2). The discharge morphology during ESE often fluctuated over time between spiky and sharp in the same individual patient. The appearance of the discharges also fluctuated between periodic and rhythmic in many patients. All patients had bilateral synchronous distribution of the discharges during ESE except one who had bilateral discharges at ESE start but developed lateralised discharges after 2 days. 3.5. Comparison between patients with unequivocal and possible ESE There was no difference in the proportion of patients with unequivocal and possible ESE having clinical convulsions (Table 1). The median NSE level at 48 h was elevated in both groups, 35 ng/ mL in the unequivocal group and 41 ng/mL in the possible group, indicating similar burden of hypoxic-ischemic injury. However, only a minority of patients had bilaterally absent SSEP N20responses, similar in both groups. Of the four survivors, three belonged to the unequivocal ESE group. We found a significant difference in EEG background continuity before the ESE start between the two groups, with continuous or nearly continuous background found more often in the possible ESE group (p = 0.029) (Table 2). There were no significant differences in background patterns after ESE start, neither for dominating background, best background or background during ESE-free periods (Table 3). The groups did not differ significantly in the amount of interictal epileptiform discharges before ESE (Table 2). There was a tendency towards higher prevalence of highly epileptiform bursts in the unequivocal group at 6 h before ESE start. 3.6. Subgroup of unequivocal ESE with evolving seizures Unequivocal ESE was defined either by repeating evolving seizures or by high frequency (P3 Hz) spike/sharp-and-waves. ESE with evolving seizures was found in 10 patients, mainly during ongoing target temperature management (Fig. 2). There were no survivors in this subgroup. These patients had a median time from CA to ESE start of 16 h (IQR 14-27) which was shorter compared to the rest of the ESE patients (42 h IQR 32-49, p = 0.001), but the prevalence of clinical convulsions (9 of 10 vs 27 of 31, p = 1.00), median NSE levels (55 ng/mL IQR 22–106 vs 36 23–61, p = 0.624) and the fraction of patients having absent N20-peaks on SSEP (2 of 6 vs 5 of 27, p = 0.584) was similar to the rest. Only one of the
10 patients with evolving seizures had recovered a continuous background prior to ESE start (1 of 10 vs 15 of 31, p = 0.059), but half of the patients had a continuous background as the dominating pattern at some point during the 72 h reviewed period after ESE start. 3.7. Routine EEG A full-montage routine EEG was performed in 35 of the 41 patients. 25 of 35 patients had ongoing possible or unequivocal ESE at this time. A generalised distribution of discharges was seen in 24 of the 25 patients, with frontal predominance in 15 of 24, the remainder having bilateral multifocal discharges. Standardised testing of EEG background reactivity was performed in 24 patients. Preserved reactivity to either sound- or pain-stimuli was seen in 8 of 13 patients in the possible ESE group and 4 of 11 in the unequivocal ESE group. 3.8. Survivors The four survivors had start of ESE after rewarming (Fig. 2). They all regained a continuous background before ESE start, without preceding highly epileptiform bursts and only occasional presence of interictal epileptiform discharges (<1/min). Their best background pattern during the following 72 h was continuous or nearly continuous. The survivors had low or moderately elevated NSE-levels at 48 h (18, 20, 25 and 37 ng/mL) compared to non-survivors (median 47, IQR 26-82 ng/mL) and all four had preserved N20-peaks. Two had clinical convulsions. Three had preserved reactivity to stimuli on the routine EEG whereas the fourth survivor was not tested for reactivity. Three patients had a good outcome at the 6-month follow-up, one with mild or no neurological deficits (CPC 1) and two with moderate disability (CPC 2). One survivor had a severe disability (CPC 3). Three survivors fulfilled the criteria for unequivocal ESE and one had possible ESE. 4. Discussion In our cohort of consecutive CA patients, ESE was a common finding affecting one third of the patients. Half of our ESE patients fulfilled our criteria for unequivocal ESE, based on the ACNS classification, but only temporarily and often during the early phase of ESE as they gradually progressed towards patterns with lower discharge frequencies. This group could not be distinguished,
Table 2 EEG characteristics before start of electrographic status epilepticus. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
P value
Highly epileptiform bursts 12 h before ESE
8/34 (24%)
5/15 (33%)
3/19 (16%)
0.417
Epileptiform discharges 12 h before ESE Abundant (P1/10 s) Frequent (1/min <1/10 s) Occasional/rare (<1/min)
12/33 (36%) 2/33 (6%) 19/33 (58%)
6/14 (43%) 1/14 (7%) 7/14 (50%)
6/19 (32%) 1/19 (5%) 12/19 (63%)
0.854
Highly epileptiform bursts 6 h before ESE
12/41 (29%)
9/22 (41%)
3/19 (16%)
0.098
Epileptiform discharges 6 h before ESE Abundant Frequent Occasional/rare
17/40 (43%) 5/40 (12%) 18/40 (45%)
9/22 (41%) 2/22 (9%) 11/22 (50%)
8/18 (44%) 3/18 (17%) 7/18 (39%)
0.748
Best background 0–4 h before ESE Continuous/nearly continuous background
16/41 (39%)
5/22 (23%)
11/19 (58%)
0.029
Occurrence of epileptiform discharges and continuous/nearly continuous background prior to ESE start. The amount of interictal epileptiform discharges was categorised according to the ACNS quantification of non-rhythmic non-periodic discharges. Data are presented as number of patients and percentages. Unequivocal and possible ESE groups are compared statistically. Numbers of patients included in the analyses are specified.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
6
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Table 3 EEG background day 1–3 after start of electrographic status epilepticus. All ESE patients (n = 41)
Unequivocal ESE group (n = 22)
Possible ESE group (n = 19)
P value
Continuous/nearly continuous as dominating background 0–24 h after ESE start 16/36 (44%) 24–48 h after ESE start 13/29 (45%) 48–72 h after ESE start 14/29 (48%)
7/19 (37%) 6/16 (38%) 7/16 (44%)
9/17 (53%) 7/13 (54%) 7/13 (54%)
0.503 0.467 0.715
Continuous/nearly continuous as best background 0–24 h after ESE start 20/37 (54%) 24–48 h after ESE start 19/30 (63%) 48–72 h after ESE start 16/29 (55%)
8/19 (42%) 8/16 (50%) 9/16 (56%)
12/18 (67%) 11/14 (79%) 7/13 (54%)
0.191 0.142 1.000
Continuous/nearly continuous as best background during ESE-free periods 0–24 h after ESE start 17/31 (55%) 24–48 h after ESE start 17/25 (68%) 48–72 h after ESE start 15/25 (60%)
6/14 (43%) 7/12 (58%) 8/13 (62%)
11/17 (65%) 10/13 (77%) 7/12 (58%)
0.289 0.411 1.000
Patients with a continuous/nearly continuous background, with or without ongoing ESE during the three 24-h periods following ESE start. The number of patients with interpretable cEEG background patterns is specified. Missing data is due to shorter monitoring periods, artefacts or persistent high discharge frequencies making the background interpretation uncertain. Unequivocal and possible ESE groups are compared statistically.
regarding clinical features including markers of brain injury, from the patients only showing possible ESE patterns. In clinical practise any pattern along the ictal-interictal continuum may or may not represent seizure activity in the individual patient. Previously used discharge frequencies for defining ESE after CA span from 0.5 Hz to 3 Hz (Hirsch et al., 2013; Amorim et al., 2015; Ruijter et al., 2015; Sivaraju and Gilmore, 2016). We chose a threshold for possible ESE in the lower range (1 Hz) to include most patterns in the continuum. The recent Salzburg Consensus Criteria for diagnosis of nonconvulsive status epilepticus proposed the term possible status epilepticus (Beniczky et al., 2013; Leitinger et al., 2016). If a possible ESE pattern occur in close relation to subtle clinical twitching it is considered diagnostic for status epilepticus. Importantly, the majority (90%) of our patients with possible ESE had clinical convulsions or myoclonus at some point after CA. A limitation with our retrospective data is that we cannot confirm temporal relationship between convulsions and EEG-discharges due to lack of video-EEG registrations. By comparing EEG characteristics of patients fulfilling our liberal definition of possible ESE with those having unequivocal ESE patterns, we aimed to investigate whether such a distinction had any bearing on prognosis. In addition, we used the serum level of NSE and the SSEP N20-responses as surrogate markers of brain injury. A correlation between NSE, EEG and other markers of postanoxic brain injury has previously been demonstrated (Cronberg et al., 2011; Rossetti et al., 2012). NSE levels were moderately or highly elevated in most patients reflecting the underlying brain injury. Interestingly, the majority of our patients in both ESE groups had present SSEP N20-peaks contradicting a very pronounced brain injury, but consistent with the notion that at least partial cortical integrity is necessary for seizure activity to occur. Development of ESE after cardiac arrest is per se a sign of poor prognosis (Rossetti et al., 2007) and the majority of these patients die, as in our cohort. An early start of ESE correlates with worse outcome (Ruijter et al., 2015). Patients who develop a continuous background prior to ESE start have been found to have a somewhat better outcome and lower NSE-levels than patients with a discontinuous or suppressed background (Rundgren et al., 2010; Dragancea et al., 2015). Our findings that the unequivocal ESE group had earlier ESE start, less often preceded by a continuous background, may indicate a more extensive brain damage. We could not support this hypothesis by any differences in other markers of brain injury such as NSE-levels, SSEP N20-potentials, presence of clinical convulsions or by their final outcome. In addition we found no differences in the EEG background continuity between the two electrographic categories once ESE had started,
but we recognise that it is difficult to define the background pattern during ongoing ESE. The subgroup of unequivocal ESE with repeating evolving seizures had similar clinical characteristics compared to the other ESE-patients. Yet, these patients had a significantly earlier start of ESE, a tendency towards a more discontinuous background prior to ESE start and none survived. Further studies with larger cohorts are needed to investigate whether this sub-categorisation has any prognostic information to add and whether it represents a distinct pathophysiological phenomenon. Although our cohort is large in comparison, the number of patients in some subgroups is small (survivors, evolving seizures). Therefore comparisons between these subgroups are limited by the lack of power and differences may have been missed. Importantly, three of our four survivors belonged to the unequivocal ESE group demonstrating that this strict definition of ESE cannot reliably identify patients with poor prognosis. None of the survivors had evolving seizures or highly epileptiform bursts. The survivors had a continuous and reactive background and were classified as unequivocal ESE based on the high discharge frequency. A similar EEG-pattern was recently described among patients surviving despite early postanoxic multifocal myoclonus (Elmer et al., 2016). In our cohort we only had four survivors with ESE, and they were all treated for a prolonged period in the ICU (16–24 days). In post cardiac arrest studies there is always a concern about the self-fulfilling prophecy. We used a multimodal protocol for prognostication at a delayed time-point. Patients with possible and unequivocal ESE were actively treated typically with increased sedation and combinations of antiepileptic agents, at least until prognostication. Whether additional patients in our cohort could have benefitted from even more aggressive and prolonged antiepileptic treatment is outside the scope of this study. We emphasize that the present study was not designed to investigate whether there are differences in the response to antiepileptic treatment among patients with unequivocal seizure patterns compared to those with possible ESE patterns. We also acknowledge that the degree of sedation may have contributed to changes in background patterns in some of our patients. Therefore we also reported the best achieved background pattern. On a group level, a recent routine EEG study showed no significant impact of ongoing sedation, in clinically used doses, on the predictive value of EEG after CA (Westhall et al., 2016). For a standard cEEG the use of at least 16 electrodes is recommended (Herman et al., 2015), but this is resource consuming and challenging to perform outside large centres. We monitored
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
our patients with a simplified EEG montage using a limited number of electrodes and acknowledge that occasional patients with seizures may be missed by this approach. The sensitivity to detect seizure activity with a simplified montage in patients with mixed aetiologies is lower (68–93%) compared to a multi-channel EEG montage (Kolls and Husain, 2007; Young et al., 2009; Karakis et al., 2010; Rubin et al., 2014). However, in adult post cardiac arrest patients the discharges most frequently show generalised distribution (Westhall et al., 2015) and a recent study on postcardiac arrest patients with simplified montage showed a substantially higher yield (92% or 100% depending on montage) (Vanherpe and Schrooten, 2016). Furthermore, the routine EEGs in our cohort often showed generalised discharges with a frontal predominance, which are most likely to be detected using our simplified montage. A reduced number of electrodes with larger inter-electrode distance may lead to slightly higher EEG amplitudes. Compared to other investigators using full-montage cEEG, we may therefore have been more prone to classify background patterns as benign (Tjepkema-Cloostermans et al., 2016). Despite these limitations our findings highlight that ESE patterns should be regarded as a continuum and that the definition of unequivocal ESE patterns based on discharge frequencies appears to be of minor importance for prognostication and outcome. Instead a multimodal prognostic approach is advocated (Callaway et al., 2015) which could also be used as a strategy for identifying those ESE patients with a potential for recovery (Rossetti et al., 2010; Westhall et al., 2013; Dragancea et al., 2015). 5. Conclusion EEG patterns compatible with status epilepticus after CA is a common finding in the ICU and have varying electrographic characteristics. Further distinction into unequivocal status epilepticus, based on the ACNS-classification, is in this study not supported by other features such as biomarkers, clinical characteristics or outcome and may therefore be of minor clinical relevance for prognostication. To support clinical decisions, a multimodal prognostication strategy incorporating other aspects of the EEG, for instance background reactivity and continuity, is advocated. Funding source Supported by the Skane University Hospital foundation, the Swedish National Health System (ALF), the County Council of Skane and the Koch Foundation. Acknowledgements The authors would like to thank Professor Ingmar Rosén (Department of Clinical Neurophysiology, Skane University Hospital, Lund, Sweden) for his valuable guidance. Conflict of interest: None of the authors have potential conflicts of interest to be disclosed. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.clinph.2017.01. 002. References Amorim E, Rittenberger JC, Baldwin ME, Callaway CW, Popescu APost Cardiac Arrest S. Malignant EEG patterns in cardiac arrest patients treated with targeted temperature management who survive to hospital discharge. Resuscitation 2015;90:127–32.
7
Beniczky S, Hirsch LJ, Kaplan PW, Pressler R, Bauer G, Aurlien H, et al. Unified EEG terminology and criteria for nonconvulsive status epilepticus. Epilepsia 2013;54 (suppl 6):28–9. Callaway CW, Soar J, Aibiki M, Bottiger BW, Brooks SC, Deakin CD, et al. Part 4: Advanced Life Support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S84–145. Chong DJ, Hirsch LJ. Which EEG patterns warrant treatment in the critically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol 2005;22:79–91. Cronberg T, Brizzi M, Liedholm LJ, Rosen I, Rubertsson S, Rylander C, et al. Neurological prognostication after cardiac arrest–recommendations from the Swedish Resuscitation Council. Resuscitation 2013;84:867–72. Cronberg T, Rundgren M, Westhall E, Englund E, Siemund R, Rosen I, et al. Neuronspecific enolase correlates with other prognostic markers after cardiac arrest. Neurology 2011;77:623–30. Dragancea I, Backman S, Westhall E, Rundgren M, Friberg H, Cronberg T. Outcome following postanoxic status epilepticus in patients with targeted temperature management after cardiac arrest. Epilepsy Behav 2015;49:173–7. Elmer J, Rittenberger JC, Faro J, Molyneaux BJ, Popescu A, Callaway CW, et al. Clinically distinct electroencephalographic phenotypes of early myoclonus after cardiac arrest. Ann Neurol 2016;80:175–84. Herman ST, Abend NS, Bleck TP, Chapman KE, Drislane FW, Emerson RG, et al. Consensus statement on continuous EEG in critically ill adults and children, part II: personnel, technical specifications, and clinical practice. J Clin Neurophysiol 2015;32:96–108. Hirsch LJ, LaRoche SM, Gaspard N, Gerard E, Svoronos A, Herman ST, et al. American clinical neurophysiology society’s standardized critical care EEG terminology: 2012 version. J Clin Neurophysiol 2013;30:1–27. Kaplan PW. EEG criteria for nonconvulsive status epilepticus. Epilepsia 2007;48 (suppl 8):39–41. Karakis I, Montouris GD, Otis JA, Douglass LM, Jonas R, Velez-Ruiz N, et al. A quick and reliable EEG montage for the detection of seizures in the critical care setting. J Clin Neurophysiol 2010;27:100–5. Kolls BJ, Husain AM. Assessment of hairline EEG as a screening tool for nonconvulsive status epilepticus. Epilepsia 2007;48:959–65. Legriel S, Hilly-Ginoux J, Resche-Rigon M, Merceron S, Pinoteau J, Henry-Lagarrigue M, et al. Prognostic value of electrographic postanoxic status epilepticus in comatose cardiac-arrest survivors in the therapeutic hypothermia era. Resuscitation 2013;84:343–50. Leitinger M, Trinka E, Gardella E, Rohracher A, Kalss G, Qerama E, et al. Diagnostic accuracy of the Salzburg EEG criteria for non-convulsive status epilepticus: a retrospective study. Lancet Neurol 2016;15:1054–62. Nolan JP, Soar J, Cariou A, Cronberg T, Moulaert VR, Deakin CD, et al. European resuscitation council and european society of intensive care medicine guidelines for post-resuscitation care 2015: Section 5 of the European Resuscitation Council Guidelines for Resuscitation 2015. Resuscitation 2015;95:202–22. Pender RA, Losey TE. A rapid course through the five electrographic stages of status epilepticus. Epilepsia 2012;53:e193–5. Rossetti AO, Carrera E, Oddo M. Early EEG correlates of neuronal injury after brain anoxia. Neurology 2012;78:796–802. Rossetti AO, Logroscino G, Liaudet L, Ruffieux C, Ribordy V, Schaller MD, et al. Status epilepticus: an independent outcome predictor after cerebral anoxia. Neurology 2007;69:255–60. Rossetti AO, Oddo M, Logroscino G, Kaplan PW. Prognostication after cardiac arrest and hypothermia: a prospective study. Ann Neurol 2010;67:301–7. Rubin MN, Jeffery OJ, Fugate JE, Britton JW, Cascino GD, Worrell GA, et al. Efficacy of a reduced electroencephalography electrode array for detection of seizures. Neurohospitalist 2014;4:6–8. Ruijter BJ, van Putten MJ, Hofmeijer J. Generalized epileptiform discharges in postanoxic encephalopathy: quantitative characterization in relation to outcome. Epilepsia 2015;56:1845–54. Rundgren M, Rosen I, Friberg H. Amplitude-integrated EEG (aEEG) predicts outcome after cardiac arrest and induced hypothermia. Intensive Care Med 2006;32:836–42. Rundgren M, Westhall E, Cronberg T, Rosen I, Friberg H. Continuous amplitudeintegrated electroencephalogram predicts outcome in hypothermia-treated cardiac arrest patients. Crit Care Med 2010;38:1838–44. Sivaraju A, Gilmore EJ. Understanding and managing the ictal-interictal continuum in neurocritical care. Curr Treat Options Neurol 2016;18:8. Tjepkema-Cloostermans MC, Hofmeijer J, Hom HW, Bosch FH, van Putten MJ. Predicting outcome in postanoxic coma: are ten EEG electrodes enough? J Clin Neurophysiol 2016. http://dx.doi.org/10.1097/WNP.0000000000000337. Treiman DM, Walton NY, Kendrick C. A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus. Epilepsy Res 1990;5:49–60. Vanherpe P, Schrooten M. Minimal EEG montage with high yield for the detection of status epilepticus in the setting of postanoxic brain damage. Acta Neurol Belg 2016. http://dx.doi.org/10.1007/s13760-016-0663-9. Westhall E, Rosen I, Rossetti AO, van Rootselaar AF, Wesenberg Kjaer T, Friberg H, et al. Interrater variability of EEG interpretation in comatose cardiac arrest patients. Clin Neurophysiol 2015;126:2397–404. Westhall E, Rossetti AO, van Rootselaar AF, Wesenberg Kjaer T, Horn J, Ullen S, et al. Standardized EEG interpretation accurately predicts prognosis after cardiac arrest. Neurology 2016;86:1482–90.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002
8
S. Backman et al. / Clinical Neurophysiology xxx (2017) xxx–xxx
Westhall E, Rundgren M, Lilja G, Friberg H, Cronberg T. Postanoxic status epilepticus can be identified and treatment guided successfully by continuous electroencephalography. Ther Hypothermia Temp Manage 2013;3:84–7. Young GB, Jordan KG, Doig GS. An assessment of nonconvulsive seizures in the intensive care unit using continuous EEG monitoring: an investigation of variables associated with mortality. Neurology 1996;47:83–9.
Young GB, Sharpe MD, Savard M, Al Thenayan E, Norton L, Davies-Schinkel C. Seizure detection with a commercially available bedside EEG monitor and the subhairline montage. Neurocrit Care 2009;11:411–6.
Please cite this article in press as: Backman S et al. Electroencephalographic characteristics of status epilepticus after cardiac arrest. Clin Neurophysiol (2017), http://dx.doi.org/10.1016/j.clinph.2017.01.002