Epilepsy & Behavior 49 (2015) 263–267
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Ictal and interictal EEG patterns in patients with nonconvulsive and subtle convulsive status epilepticus Tushar Divakar Gosavi ⁎, Siew Ju See, Shih Hui Lim National Neuroscience Institute and Singapore General Hospital, Singapore Dukes-NUS Medical School, Singapore
a r t i c l e
i n f o
Article history: Revised 5 May 2015 Accepted 6 May 2015 Available online 14 June 2015 Keywords: Status epilepticus Nonconvulsive EEG
a b s t r a c t Background: Electroencephalography findings in nonconvulsive or subtle convulsive status epilepticus (NCSE and SCSE, respectively) can be heterogenous. We aimed to study the different patterns on EEG in our cohort of patients. Objective: Our objective was to study ictal and interictal EEG patterns in patients with NCSE and SCSE. Methods: From January 2012 to December 2013, EEGs recorded from patients admitted for altered mental status suspected of having NCSE or SCSE were reviewed retrospectively. Electroencephalography status was defined as having (a) continuous ictal discharges lasting N5 min or (b) N 2 discrete bursts of ictal discharges, each lasting b 5 min, without returning to previous background rhythm in between these bursts. Results: Among 1698 EEGs recorded for at least 30 min from hospitalized patients, 55 (3.23%) satisfied the criteria of EEG SE. The ictal onset was regional in 37 (67.2%) EEGs, multiregional independent in 8 (14.5%), and generalized in 10 (18.4%). The EEG seizure duration was N 5 min in 24 (43.6%) EEGs, between 1 and 5 min in 14 (25.4%), and less than 1 min in 17 (30.8%). Twenty (36.3%) EEGs showed one continuous prolonged seizure episode of N5-minute duration, 15 (27.2%) had 10 or less discrete episodes, 20 (36.3%) had more than 10 episodes, and 11 (20%) had 2 or more ictal patterns. Thirty (54.5%) EEGs had onset ictal frequency of N8Hz whereas the rest had b 8-Hz ictal frequency. In the interictal segment, 29 patients had continuous generalized slow waves, while 12 had intermittent generalized slow waves. Eleven patients had continuous slow waves lateralized to one hemisphere, and these were ipsilateral to the ictal focus in 10 but contralateral in 1. Other interictal waves seen were PLEDS (6), sharp waves (3), suppression (5), and triphasic waves (1). The background alpha rhythm was absent in 36 patients and slow in 14, and normal background alpha was seen in the interictal period in 5 patients. Conclusion: The ictal and interictal EEG patterns in NCSE and SCSE can be varied. Further study to look for etiologic and clinical correlates of each pattern could add to its clinical value. This article is part of a Special Issue entitled “Status Epilepticus”. © 2015 Elsevier Inc. All rights reserved.
1. Background Nonconvulsive or subtle convulsive EEG status epilepticus is an important clinical entity and is suspected in patients who have an unexplained altered mentation. The clinical manifestations can be varied and can range from mere drowsiness to subtle muscle twitches. It is often difficult to diagnose clinically, and an EEG clinches the diagnosis.
Electroencephalography patterns in nonconvulsive status epilepticus (NCSE) are diverse and different from those seen in typical convulsive seizures. They are not extensively studied. There is uncertainty as to whether a particular pattern suggests an underlying specific etiology or whether it could be of any prognostic significance. As a first step to systematically study this, we decided to analyze the different types of EEG patterns in our cohort of patients with NCSE. 2. Aim
⁎ Corresponding author at: Dept. of Neurology, National Neuroscience Institute, Jalan Tan Tock Seng, Singapore 308433. Tel.: +65 98164637. E-mail addresses: [email protected] (T.D. Gosavi), [email protected] (S.J. See), [email protected] (S.H. Lim).
http://dx.doi.org/10.1016/j.yebeh.2015.05.011 1525-5050/© 2015 Elsevier Inc. All rights reserved.
Our aim was to study the EEG patterns in all patients diagnosed with NCSE over a period of two years from January 2012 to December 2013.
264
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267 Regional
3. Methodology 0
All the patients with a diagnosis of EEG status during a period of two years between January 2012 and December 2013 were identified from the database of the EEG laboratory at the Singapore General Hospital. Among these, patients who had obvious clinical GTC seizures or simple and complex partial seizures during the 30-minute EEG recording were excluded. Only those patients with nonconvulsive status epilepticus or those with subtle involuntary movements were included for final analysis. All the EEGs were recorded with electrodes placed according to the international 10–20 system. They were performed with one of the two Compumedics EEG machines in our laboratory. The EEGs were reported by neurologists trained in EEG and epilepsy. The following criteria were used to diagnose nonconvulsive EEG status epilepticus: (a) Continuous ictal discharges lasting N5 min, (b) N 2 discrete bursts of ictal discharges, each lasting b5 min, without returning to previous background rhythm in between these bursts, or (c) combination of a and b associated with a change in mental state from the patient's baseline premorbid mental state with or without subtle involuntary motor activity. To define NCSE, we included all epileptiform discharges as ictal irrespective of their frequency (whether b or N2.5 Hz) if there was a definite spatiotemporal evolution. We also included paroxysmal fast activity as ictal if there was a definite evolution. The discharges had to be either continuous or recurrent during the 30-minute EEG for inclusion. The EEGs showing NCSE based on the above criteria were reviewed for the ictal and interictal patterns. Many patients had serial EEG SE. For these patients, the first EEG that made the diagnosis was used for the purpose of analysis. The following ictal parameters were taken into account: 1) 2) 3) 4)
Distribution of seizures The frequency of ictal discharges The amplitude of ictal discharges The persistence of ictal discharges
Based on the distribution of the ictal discharge, the EEGs were classified as either regional (focal), multiregional independent (multiple independent foci), or generalized (with no obvious focality or lateralization). The frequency of ictal discharge was determined by visual analysis. It was defined as the predominant ictal frequency within the first 10-second page of the first seizure seen on EEG. They were then classified as slow (frequency b 8 Hz) and fast (frequency N 8 Hz) activities. The amplitude of ictal discharges was classified as low (b20 μV), medium (20–70 μV), and high (N 70 μV). The persistence was classified as either continuous single ictal discharge lasting more than 5 min or as multiple discrete episodes with duration of each seizure lasting b 5 min. The length of the longest seizure and number of seizures were noted in cases where EEGs showed multiple episodes. The data collected was the component of a larger study of nonconvulsive status epilepticus approved by the institutional review board. 4. Results Among 1698 EEGs recorded for at least 30 min from hospitalized patients, 55 (3.23%) satisfied the above criteria of EEG SE. The ictal onset was regional in 37 (67.2%) EEGs, multiregional independent in 8 (14.5%), and generalized in 10 (18.4%) (Fig. 1). Among the patients with focal onset, only two patients had secondary generalization seen on the EEG.
Multiregional independent
18.4%
Generalised
14.5% 67.2%
Fig. 1. Proportion of NCSE by focality.
Of the 37 SE cases with regional onset, 9 had diffuse regional or hemispheric onset, and the remaining 28 cases had a more focal regional onset. Among these 28 cases, 12 had temporal involvement at the onset (4 — frontotemporal onset, 4—temporal onset, and 4—temporoparietal– occipital) while in 16 cases, there was pure extratemporal involvement: frontocentral (8 cases), centroparietal (4 cases), and parietooccipital (4 cases). Thirty (54.5%) EEGs had a fast ictal onset frequency of ≥8Hz whereas the rest had b8Hz ictal onset frequency (Fig. 2A, B, and C). The subgroup of patients with regional discharges was compared to those with generalized discharges with respect to the ictal onset frequencies. Electroencephalograms with regional discharges were more likely to have faster onset frequencies than those with generalized discharges (OR = 5.7, 95% CI = 1.04 to 31.67). When multiregional independent discharges were combined together with the pure regional group and compared with the generalized group, the association with faster frequencies would become weaker (OR = 4.78, 95% CI = 0.89 to 25.67). The ictal onset discharge amplitude was low to medium in 58.1% of EEGs while the rest had medium to high amplitude ictal onset discharge. Low amplitude ictal onset discharge was associated with higher frequencies in the respective EEGs (OR = 4.14, 95% CI = 1.28 to 13.31) and vice versa. Also, focal seizures were more likely to have a low to medium amplitude ictal discharge whereas generalized distribution correlated more with medium to high amplitude discharges (OR = 4.22, 95% CI = 0.95 to 18.65). The longest EEG seizure duration was N 5 min in 24 (43.6%) EEGs, between 1 and 5 min in 14 (25.4%), and less than 1 min in 17 (30.8%). Analysis of the duration of longest seizure compared with seizure distribution is shown in Table 1. There was no statistically significant relation between these. Twenty (36.3%) EEGs showed one continuous prolonged seizure episode of N5-minute duration, 15 (27.2%) had 10 or less discrete episodes, 20 (36.3%) had more than 10 episodes, and 11 (20%) had 2 or more ictal patterns. In the interictal segment, 29 patients had continuous generalized slow waves, while 12 had intermittent generalized slow waves. Eleven patients had continuous slow waves lateralized to one hemisphere, and these were ipsilateral to the ictal focus in 10 but contralateral in 1. Other interictal waves seen were PLEDS (6), sharp waves (3), suppression (5), and triphasic waves (1). The expected background alpha rhythm was absent in 36 patients and slow in 14, and normal background alpha was seen in the interictal period in 5 patients. 5. Discussion The frequency of NCSE differs depending on the denominator population. In previous studies, NCSE was observed in 8% of comatose patients in an ICU [1], in 5% of all patients with altered mental status attending emergency departments [2], in 9.3% inpatients referred for EEG with clinical suspicion of NCSE [3], and in 9.8% of inpatient emergent EEGs [4]. The frequency in our study was lower compared to the above numbers as our denominator was all the inpatient EEGs. The indications for EEG in our cohort were heterogeneous with the patients
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
265
Fig. 2. EEG NCSE with fast frequency (A and B); EEG NCSE with slow frequency (C).
being a mix of comatose ICU cases, inpatients with suspicion of NCSE, as well as relatively stable inpatients admitted with a history of an episode of loss of consciousness. This could have diluted the severity of indication in our cohort. However, our frequency of 3.23% was similar to the study by Haffey et al., whose denominator cohort was similar to ours, namely, all the inpatient EEGs [5]. Our criteria for diagnosis of NCSE were similar to those suggested by Benickzy et al. [6] and Young et al. [7] except that we did not use the response to treatment criteria and that we did include paroxysmal fast activity as ictal if there was a definite temporospatial evolution. This being a retrospective study, with patients being treated by multiple physicians and neurologists, antiepileptics were not consistently administered during the EEG itself for us to assess the response. However, we did take into account the change in clinical state from patients' premorbid baseline as a part of inclusion criteria. Also, we did refrain
Table 1 Duration of longest seizure according to seizure distribution. Seizure distribution
Regional Multiregional independent Generalized
Duration of longest seizure b10 s
10 to 30 s
31 to 60 s
N1 to 5 min
N5 min
6 0 2
1 1 2
5 0 0
10 3 1
15 4 5
from calling GPEDS, PLEDS, triphasic waves, and burst suppression patterns as seizures unless there was a definite evolution, thus, minimizing the possibility of over diagnosis. A further clinical–electrographic classification of NCSE into that of simple partial SE, complex partial SE, and NCSE with coma which in turn can be further subdivided into subtle convulsive generalized SE and coma induced by CNS disorders associated with SE could be useful [8]. However, we did not electroclinically classify our patients further but rather, classified all the patients as having NCSE. This was so because we wanted to study the diversity of EEG patterns encountered, due to which it was imperative for us to include all the cases with diverse etiologies that satisfied our EEG criteria. Hence, the spectrum of our patients ranged from conscious patients with focal discharges to comatose patients with hypoxic–ischemic encephalopathy, provided they had a definite change in mental state from their known baseline. Regional seizures were the commonest. Most of the patients with regional onset continued to have pure regional seizures. Only two had secondary generalization, both of which had temporal lobe onset. Among the regional seizures, pure regional involvement was thrice more frequent than hemispheric involvement. Categorization of ictal EEG patterns for NCSE might be different from the conventional ictal EEG patterns associated with GTCSE or CPSE. These have not been extensively studied in the past. A few studies have looked at these patterns. Granner and Soo [9] reviewed 85 ictal episodes and noted that the morphologies and
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
Fig. 3. Typical evolution of NCSE.
266
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
patterns of persistence greatly varied as did the frequency of ictal discharges. They studied the predominant ictal frequency in three different groups of patients and found that the mean predominant ictal frequency was fairly uniform in patients who had purely generalized discharges, generalized discharges with focal predominance, and clear focal discharges, and it ranged between 1.0 and 3.5 Hz. We strictly observed the ictal onset frequency to minimize the complexity and subjectivity in establishing the predominant frequency throughout the status. Although Granner and Soo [9] noted that 69% of patients had predominantly generalized discharges (with another 13% showing bilateral widely distributed discharges with focal predominance), in another study of NCSE in pediatric population [10], focal EEG seizures were seen in more than three-quarters of patients, and these seizures had a mean frequency of 3.5 Hz; for those with a generalized ictal discharges which constituted less than a quarter of cases, mean frequency was 1.9 Hz. In our study, we did have similar findings to the latter, with respect to seizure distribution: regional seizures being the commonest with 67% of the cases (with another 14% with multiregional distribution). However, the ictal onset frequency in our series was evenly distributed across the cutoff of 8 Hz. The predominance of faster frequency seen in our group of patients as opposed to other previous observations could be largely due to the fact that our frequencies were strictly the first 10 s of the first seizure and we did not look into the predominant frequency throughout the entire status which could be very tedious to do if the results of the study were to be applied in everyday clinical practice. With respect to the morphology, our findings did not suggest the occurrence of typical ictal rhythms like polyspikes or spike-and-wave discharges. Rather, the pattern observed with high frequency discharges was that of a sudden appearance of a low to medium voltage paroxysmal fast activity obviously different from the preictal EEG, followed by its persistence for variable lengths of time and disappearance such that the preictal EEG pattern would reappear (Fig. 3). This was different compared to the prototype EEG evolution seen in phases of a GTC seizure and could reflect the fundamental difference in the brain electric activities in GTCS and NCS. Our categorization of frequencies and amplitudes was rather arbitrary but was essential to systematically study the EEG patterns. The amplitude definitions were adopted from our EEG laboratory standards. The frequency cutoff of b8 Hz to classify frequency as slow was based on the concept of subalpha frequencies as slow. For some seizures with slow ictal onset frequencies, the distinction between an encephalopathic EEG and NCSE EEG could be difficult at times, and there would be observer bias and differences in opinion among neurologists. This is especially true when there is no clinical manifestation. However, it is a well accepted fact that the evolution in time and space of a discharge is helpful in differentiating the EEG of encephalopathies from that of NCSE [11]. 6. Limitations of the study Being a retrospective study, it has the limitations associated with such analyses. The EEGs were not continuous but rather 30-minute EEGs. In addition, some patients could have had the NCSE ongoing for some time prior to the first EEG, and many would have had the status for variable durations after the EEG. Although continuous EEG would be the best
267
way to demonstrate the duration of status, number of seizures, and length of the longest seizure, it is often not easily available and is costly. Although periodic discharges with a lower frequency (b 3 Hz) without a definite evolution were not considered status, a lack of exact definition of evolution could have led to overdiagnosis of NCSE among some patients with predominantly slower ictal onset frequency that had only subtle evolution. Also, we did not use the “response to treatment” factor to diagnose NCSE. Although an electro clinical response to antiepileptics could be a helpful diagnostic process, nevertheless, an absence of such response does not exclude diagnosis. 7. Clinical significance Whether the findings of EEG patterns could have a clinical implication would be interesting to study. With this question in mind, one would consider a larger study with the clinical aspect involved to see if any particular pattern of NCSE has a clinical, etiologic, or prognostic association like that of triphasic waves with metabolic or toxic encephalopathy. Such a study would certainly need a larger number of patients. 8. Conclusion Electroencephalography patterns in NCSE could be varied and importantly different from the patterns associated with typical GTCS or absence seizures. This could suggest NCSE as a unique entity electroclinically different from convulsive status epilepticus. Conflicts of interest None of the authors have any conflicts of interest. References [1] Towne AR, Waterhouse EJ, Boggs JG, Garnett LK, Brown AJ, Smith JR, et al. Prevalence of nonconvulsive status epilepticus in comatose patients. Neurology 2000;54:340–5. [2] Zehtabchi S, Baki SGA, Omurtag A, Sinert R, Chari G, Malhotra S, et al. Prevalence of non-convulsive seizure and other electroencephalographic abnormalities in emergency department patients with altered mental status. Am J Emerg Med 2013;31. http://dx.doi.org/10.1016/ j.ajem. 2013.08.002. [3] Alroughani R, Javidan M, Qasem A, Alotaibi N. Non-convulsive status epilepticus; the rate of occurrence in a general hospital. Seizure 2009;18:38–42. [4] Tu TM, Loh NK, Tan NCK. Clinical risk factors for non-convulsive status epilepticus during emergent electroencephalogram. Seizure 2013;22:794–7. [5] Haffey S, McKernan A, Pang K. Non-convulsive status epilepticus: a profile of patients diagnosed within a tertiary referral centre. J Neurol Neurosurg Psychiatry 2004;75:1043–4. [6] 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. [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. [8] Meierkord H, Holtkamp M. Non-convulsive status epilepticus in adults: clinical forms and treatment. Lancet Neurol 2007;6:329–39. [9] Granner MA, Soo IL. Nonconvulsive status epilepticus: EEG analysis in a large series. Epilepsia 1994;35(1):42–7. [10] Tay SKH, Hirsh LJ, Leary L, Jette N, Wittman J, Akman CI. Nonconvulsive status epilepticus in children: clinical and EEG characteristics. Epilepsia 2006;47(9): 1504–9. [11] Treiman DM, Walker MC. Treatment of seizure emergencies: convulsive and nonconvulsive status epilepticus. Epilepsy Res 2006;68(Suppl. 1):S77–82.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Ictal and interictal EEG patterns in patients with nonconvulsive and subtle convulsive status epilepticus Tushar Divakar Gosavi ⁎, Siew Ju See, Shih Hui Lim National Neuroscience Institute and Singapore General Hospital, Singapore Dukes-NUS Medical School, Singapore
a r t i c l e
i n f o
Article history: Revised 5 May 2015 Accepted 6 May 2015 Available online 14 June 2015 Keywords: Status epilepticus Nonconvulsive EEG
a b s t r a c t Background: Electroencephalography findings in nonconvulsive or subtle convulsive status epilepticus (NCSE and SCSE, respectively) can be heterogenous. We aimed to study the different patterns on EEG in our cohort of patients. Objective: Our objective was to study ictal and interictal EEG patterns in patients with NCSE and SCSE. Methods: From January 2012 to December 2013, EEGs recorded from patients admitted for altered mental status suspected of having NCSE or SCSE were reviewed retrospectively. Electroencephalography status was defined as having (a) continuous ictal discharges lasting N5 min or (b) N 2 discrete bursts of ictal discharges, each lasting b 5 min, without returning to previous background rhythm in between these bursts. Results: Among 1698 EEGs recorded for at least 30 min from hospitalized patients, 55 (3.23%) satisfied the criteria of EEG SE. The ictal onset was regional in 37 (67.2%) EEGs, multiregional independent in 8 (14.5%), and generalized in 10 (18.4%). The EEG seizure duration was N 5 min in 24 (43.6%) EEGs, between 1 and 5 min in 14 (25.4%), and less than 1 min in 17 (30.8%). Twenty (36.3%) EEGs showed one continuous prolonged seizure episode of N5-minute duration, 15 (27.2%) had 10 or less discrete episodes, 20 (36.3%) had more than 10 episodes, and 11 (20%) had 2 or more ictal patterns. Thirty (54.5%) EEGs had onset ictal frequency of N8Hz whereas the rest had b 8-Hz ictal frequency. In the interictal segment, 29 patients had continuous generalized slow waves, while 12 had intermittent generalized slow waves. Eleven patients had continuous slow waves lateralized to one hemisphere, and these were ipsilateral to the ictal focus in 10 but contralateral in 1. Other interictal waves seen were PLEDS (6), sharp waves (3), suppression (5), and triphasic waves (1). The background alpha rhythm was absent in 36 patients and slow in 14, and normal background alpha was seen in the interictal period in 5 patients. Conclusion: The ictal and interictal EEG patterns in NCSE and SCSE can be varied. Further study to look for etiologic and clinical correlates of each pattern could add to its clinical value. This article is part of a Special Issue entitled “Status Epilepticus”. © 2015 Elsevier Inc. All rights reserved.
1. Background Nonconvulsive or subtle convulsive EEG status epilepticus is an important clinical entity and is suspected in patients who have an unexplained altered mentation. The clinical manifestations can be varied and can range from mere drowsiness to subtle muscle twitches. It is often difficult to diagnose clinically, and an EEG clinches the diagnosis.
Electroencephalography patterns in nonconvulsive status epilepticus (NCSE) are diverse and different from those seen in typical convulsive seizures. They are not extensively studied. There is uncertainty as to whether a particular pattern suggests an underlying specific etiology or whether it could be of any prognostic significance. As a first step to systematically study this, we decided to analyze the different types of EEG patterns in our cohort of patients with NCSE. 2. Aim
⁎ Corresponding author at: Dept. of Neurology, National Neuroscience Institute, Jalan Tan Tock Seng, Singapore 308433. Tel.: +65 98164637. E-mail addresses: [email protected] (T.D. Gosavi), [email protected] (S.J. See), [email protected] (S.H. Lim).
http://dx.doi.org/10.1016/j.yebeh.2015.05.011 1525-5050/© 2015 Elsevier Inc. All rights reserved.
Our aim was to study the EEG patterns in all patients diagnosed with NCSE over a period of two years from January 2012 to December 2013.
264
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267 Regional
3. Methodology 0
All the patients with a diagnosis of EEG status during a period of two years between January 2012 and December 2013 were identified from the database of the EEG laboratory at the Singapore General Hospital. Among these, patients who had obvious clinical GTC seizures or simple and complex partial seizures during the 30-minute EEG recording were excluded. Only those patients with nonconvulsive status epilepticus or those with subtle involuntary movements were included for final analysis. All the EEGs were recorded with electrodes placed according to the international 10–20 system. They were performed with one of the two Compumedics EEG machines in our laboratory. The EEGs were reported by neurologists trained in EEG and epilepsy. The following criteria were used to diagnose nonconvulsive EEG status epilepticus: (a) Continuous ictal discharges lasting N5 min, (b) N 2 discrete bursts of ictal discharges, each lasting b5 min, without returning to previous background rhythm in between these bursts, or (c) combination of a and b associated with a change in mental state from the patient's baseline premorbid mental state with or without subtle involuntary motor activity. To define NCSE, we included all epileptiform discharges as ictal irrespective of their frequency (whether b or N2.5 Hz) if there was a definite spatiotemporal evolution. We also included paroxysmal fast activity as ictal if there was a definite evolution. The discharges had to be either continuous or recurrent during the 30-minute EEG for inclusion. The EEGs showing NCSE based on the above criteria were reviewed for the ictal and interictal patterns. Many patients had serial EEG SE. For these patients, the first EEG that made the diagnosis was used for the purpose of analysis. The following ictal parameters were taken into account: 1) 2) 3) 4)
Distribution of seizures The frequency of ictal discharges The amplitude of ictal discharges The persistence of ictal discharges
Based on the distribution of the ictal discharge, the EEGs were classified as either regional (focal), multiregional independent (multiple independent foci), or generalized (with no obvious focality or lateralization). The frequency of ictal discharge was determined by visual analysis. It was defined as the predominant ictal frequency within the first 10-second page of the first seizure seen on EEG. They were then classified as slow (frequency b 8 Hz) and fast (frequency N 8 Hz) activities. The amplitude of ictal discharges was classified as low (b20 μV), medium (20–70 μV), and high (N 70 μV). The persistence was classified as either continuous single ictal discharge lasting more than 5 min or as multiple discrete episodes with duration of each seizure lasting b 5 min. The length of the longest seizure and number of seizures were noted in cases where EEGs showed multiple episodes. The data collected was the component of a larger study of nonconvulsive status epilepticus approved by the institutional review board. 4. Results Among 1698 EEGs recorded for at least 30 min from hospitalized patients, 55 (3.23%) satisfied the above criteria of EEG SE. The ictal onset was regional in 37 (67.2%) EEGs, multiregional independent in 8 (14.5%), and generalized in 10 (18.4%) (Fig. 1). Among the patients with focal onset, only two patients had secondary generalization seen on the EEG.
Multiregional independent
18.4%
Generalised
14.5% 67.2%
Fig. 1. Proportion of NCSE by focality.
Of the 37 SE cases with regional onset, 9 had diffuse regional or hemispheric onset, and the remaining 28 cases had a more focal regional onset. Among these 28 cases, 12 had temporal involvement at the onset (4 — frontotemporal onset, 4—temporal onset, and 4—temporoparietal– occipital) while in 16 cases, there was pure extratemporal involvement: frontocentral (8 cases), centroparietal (4 cases), and parietooccipital (4 cases). Thirty (54.5%) EEGs had a fast ictal onset frequency of ≥8Hz whereas the rest had b8Hz ictal onset frequency (Fig. 2A, B, and C). The subgroup of patients with regional discharges was compared to those with generalized discharges with respect to the ictal onset frequencies. Electroencephalograms with regional discharges were more likely to have faster onset frequencies than those with generalized discharges (OR = 5.7, 95% CI = 1.04 to 31.67). When multiregional independent discharges were combined together with the pure regional group and compared with the generalized group, the association with faster frequencies would become weaker (OR = 4.78, 95% CI = 0.89 to 25.67). The ictal onset discharge amplitude was low to medium in 58.1% of EEGs while the rest had medium to high amplitude ictal onset discharge. Low amplitude ictal onset discharge was associated with higher frequencies in the respective EEGs (OR = 4.14, 95% CI = 1.28 to 13.31) and vice versa. Also, focal seizures were more likely to have a low to medium amplitude ictal discharge whereas generalized distribution correlated more with medium to high amplitude discharges (OR = 4.22, 95% CI = 0.95 to 18.65). The longest EEG seizure duration was N 5 min in 24 (43.6%) EEGs, between 1 and 5 min in 14 (25.4%), and less than 1 min in 17 (30.8%). Analysis of the duration of longest seizure compared with seizure distribution is shown in Table 1. There was no statistically significant relation between these. Twenty (36.3%) EEGs showed one continuous prolonged seizure episode of N5-minute duration, 15 (27.2%) had 10 or less discrete episodes, 20 (36.3%) had more than 10 episodes, and 11 (20%) had 2 or more ictal patterns. In the interictal segment, 29 patients had continuous generalized slow waves, while 12 had intermittent generalized slow waves. Eleven patients had continuous slow waves lateralized to one hemisphere, and these were ipsilateral to the ictal focus in 10 but contralateral in 1. Other interictal waves seen were PLEDS (6), sharp waves (3), suppression (5), and triphasic waves (1). The expected background alpha rhythm was absent in 36 patients and slow in 14, and normal background alpha was seen in the interictal period in 5 patients. 5. Discussion The frequency of NCSE differs depending on the denominator population. In previous studies, NCSE was observed in 8% of comatose patients in an ICU [1], in 5% of all patients with altered mental status attending emergency departments [2], in 9.3% inpatients referred for EEG with clinical suspicion of NCSE [3], and in 9.8% of inpatient emergent EEGs [4]. The frequency in our study was lower compared to the above numbers as our denominator was all the inpatient EEGs. The indications for EEG in our cohort were heterogeneous with the patients
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
265
Fig. 2. EEG NCSE with fast frequency (A and B); EEG NCSE with slow frequency (C).
being a mix of comatose ICU cases, inpatients with suspicion of NCSE, as well as relatively stable inpatients admitted with a history of an episode of loss of consciousness. This could have diluted the severity of indication in our cohort. However, our frequency of 3.23% was similar to the study by Haffey et al., whose denominator cohort was similar to ours, namely, all the inpatient EEGs [5]. Our criteria for diagnosis of NCSE were similar to those suggested by Benickzy et al. [6] and Young et al. [7] except that we did not use the response to treatment criteria and that we did include paroxysmal fast activity as ictal if there was a definite temporospatial evolution. This being a retrospective study, with patients being treated by multiple physicians and neurologists, antiepileptics were not consistently administered during the EEG itself for us to assess the response. However, we did take into account the change in clinical state from patients' premorbid baseline as a part of inclusion criteria. Also, we did refrain
Table 1 Duration of longest seizure according to seizure distribution. Seizure distribution
Regional Multiregional independent Generalized
Duration of longest seizure b10 s
10 to 30 s
31 to 60 s
N1 to 5 min
N5 min
6 0 2
1 1 2
5 0 0
10 3 1
15 4 5
from calling GPEDS, PLEDS, triphasic waves, and burst suppression patterns as seizures unless there was a definite evolution, thus, minimizing the possibility of over diagnosis. A further clinical–electrographic classification of NCSE into that of simple partial SE, complex partial SE, and NCSE with coma which in turn can be further subdivided into subtle convulsive generalized SE and coma induced by CNS disorders associated with SE could be useful [8]. However, we did not electroclinically classify our patients further but rather, classified all the patients as having NCSE. This was so because we wanted to study the diversity of EEG patterns encountered, due to which it was imperative for us to include all the cases with diverse etiologies that satisfied our EEG criteria. Hence, the spectrum of our patients ranged from conscious patients with focal discharges to comatose patients with hypoxic–ischemic encephalopathy, provided they had a definite change in mental state from their known baseline. Regional seizures were the commonest. Most of the patients with regional onset continued to have pure regional seizures. Only two had secondary generalization, both of which had temporal lobe onset. Among the regional seizures, pure regional involvement was thrice more frequent than hemispheric involvement. Categorization of ictal EEG patterns for NCSE might be different from the conventional ictal EEG patterns associated with GTCSE or CPSE. These have not been extensively studied in the past. A few studies have looked at these patterns. Granner and Soo [9] reviewed 85 ictal episodes and noted that the morphologies and
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
Fig. 3. Typical evolution of NCSE.
266
T.D. Gosavi et al. / Epilepsy & Behavior 49 (2015) 263–267
patterns of persistence greatly varied as did the frequency of ictal discharges. They studied the predominant ictal frequency in three different groups of patients and found that the mean predominant ictal frequency was fairly uniform in patients who had purely generalized discharges, generalized discharges with focal predominance, and clear focal discharges, and it ranged between 1.0 and 3.5 Hz. We strictly observed the ictal onset frequency to minimize the complexity and subjectivity in establishing the predominant frequency throughout the status. Although Granner and Soo [9] noted that 69% of patients had predominantly generalized discharges (with another 13% showing bilateral widely distributed discharges with focal predominance), in another study of NCSE in pediatric population [10], focal EEG seizures were seen in more than three-quarters of patients, and these seizures had a mean frequency of 3.5 Hz; for those with a generalized ictal discharges which constituted less than a quarter of cases, mean frequency was 1.9 Hz. In our study, we did have similar findings to the latter, with respect to seizure distribution: regional seizures being the commonest with 67% of the cases (with another 14% with multiregional distribution). However, the ictal onset frequency in our series was evenly distributed across the cutoff of 8 Hz. The predominance of faster frequency seen in our group of patients as opposed to other previous observations could be largely due to the fact that our frequencies were strictly the first 10 s of the first seizure and we did not look into the predominant frequency throughout the entire status which could be very tedious to do if the results of the study were to be applied in everyday clinical practice. With respect to the morphology, our findings did not suggest the occurrence of typical ictal rhythms like polyspikes or spike-and-wave discharges. Rather, the pattern observed with high frequency discharges was that of a sudden appearance of a low to medium voltage paroxysmal fast activity obviously different from the preictal EEG, followed by its persistence for variable lengths of time and disappearance such that the preictal EEG pattern would reappear (Fig. 3). This was different compared to the prototype EEG evolution seen in phases of a GTC seizure and could reflect the fundamental difference in the brain electric activities in GTCS and NCS. Our categorization of frequencies and amplitudes was rather arbitrary but was essential to systematically study the EEG patterns. The amplitude definitions were adopted from our EEG laboratory standards. The frequency cutoff of b8 Hz to classify frequency as slow was based on the concept of subalpha frequencies as slow. For some seizures with slow ictal onset frequencies, the distinction between an encephalopathic EEG and NCSE EEG could be difficult at times, and there would be observer bias and differences in opinion among neurologists. This is especially true when there is no clinical manifestation. However, it is a well accepted fact that the evolution in time and space of a discharge is helpful in differentiating the EEG of encephalopathies from that of NCSE [11]. 6. Limitations of the study Being a retrospective study, it has the limitations associated with such analyses. The EEGs were not continuous but rather 30-minute EEGs. In addition, some patients could have had the NCSE ongoing for some time prior to the first EEG, and many would have had the status for variable durations after the EEG. Although continuous EEG would be the best
267
way to demonstrate the duration of status, number of seizures, and length of the longest seizure, it is often not easily available and is costly. Although periodic discharges with a lower frequency (b 3 Hz) without a definite evolution were not considered status, a lack of exact definition of evolution could have led to overdiagnosis of NCSE among some patients with predominantly slower ictal onset frequency that had only subtle evolution. Also, we did not use the “response to treatment” factor to diagnose NCSE. Although an electro clinical response to antiepileptics could be a helpful diagnostic process, nevertheless, an absence of such response does not exclude diagnosis. 7. Clinical significance Whether the findings of EEG patterns could have a clinical implication would be interesting to study. With this question in mind, one would consider a larger study with the clinical aspect involved to see if any particular pattern of NCSE has a clinical, etiologic, or prognostic association like that of triphasic waves with metabolic or toxic encephalopathy. Such a study would certainly need a larger number of patients. 8. Conclusion Electroencephalography patterns in NCSE could be varied and importantly different from the patterns associated with typical GTCS or absence seizures. This could suggest NCSE as a unique entity electroclinically different from convulsive status epilepticus. Conflicts of interest None of the authors have any conflicts of interest. References [1] Towne AR, Waterhouse EJ, Boggs JG, Garnett LK, Brown AJ, Smith JR, et al. Prevalence of nonconvulsive status epilepticus in comatose patients. Neurology 2000;54:340–5. [2] Zehtabchi S, Baki SGA, Omurtag A, Sinert R, Chari G, Malhotra S, et al. Prevalence of non-convulsive seizure and other electroencephalographic abnormalities in emergency department patients with altered mental status. Am J Emerg Med 2013;31. http://dx.doi.org/10.1016/ j.ajem. 2013.08.002. [3] Alroughani R, Javidan M, Qasem A, Alotaibi N. Non-convulsive status epilepticus; the rate of occurrence in a general hospital. Seizure 2009;18:38–42. [4] Tu TM, Loh NK, Tan NCK. Clinical risk factors for non-convulsive status epilepticus during emergent electroencephalogram. Seizure 2013;22:794–7. [5] Haffey S, McKernan A, Pang K. Non-convulsive status epilepticus: a profile of patients diagnosed within a tertiary referral centre. J Neurol Neurosurg Psychiatry 2004;75:1043–4. [6] 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. [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. [8] Meierkord H, Holtkamp M. Non-convulsive status epilepticus in adults: clinical forms and treatment. Lancet Neurol 2007;6:329–39. [9] Granner MA, Soo IL. Nonconvulsive status epilepticus: EEG analysis in a large series. Epilepsia 1994;35(1):42–7. [10] Tay SKH, Hirsh LJ, Leary L, Jette N, Wittman J, Akman CI. Nonconvulsive status epilepticus in children: clinical and EEG characteristics. Epilepsia 2006;47(9): 1504–9. [11] Treiman DM, Walker MC. Treatment of seizure emergencies: convulsive and nonconvulsive status epilepticus. Epilepsy Res 2006;68(Suppl. 1):S77–82.
