Nonconvulsive Status Epilepticus in Metabolic Encephalopathy in Indonesia Referral Hospital
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.314533
Source of Support: None, Conflict of Interest: None
Keywords: Metabolic encephalopathy, mortality, non-convulsive status epilepticusKey Messages: Treating the underlying etiology of nonconvulsive status epilepticus was important to decrease the mortality rate; aggressive anti-epileptic treatment should be carefully considered since AED may contribute to morbidity and mortality of critically ill patients.
Nonconvulsive status epilepticus (NCSE) is an epileptic seizure without prominent motoric features. Electroencephalography (EEG) plays a vital role in the diagnosis of NCSE. The EEG activity of NCSE varied widely and some diagnostic criteria had been proposed., The Salzburg EEG criteria for NCSE was known to have high accuracy.,
The mortality of NCSE patients was 7.7% to 57% and affected by the underlying etiology.,, NCSE was found in the critically ill populations with various aetiologies, including metabolic encephalopathy (ME). Delay in diagnosis of NCSE might contribute to poor outcome., Patients with sepsis had a higher rate of non-convulsive seizures and poorer outcomes. Seizure and NCSE could be the initial symptoms of ME., Clinical evidence of NCSE includes agitation, delirium, altered mental status without obvious convulsive seizure. This study aimed to investigate the incidence and outcome of NCSE in ME at our hospital as a referral hospital in Indonesia.
This was a cross-sectional prospective study in Cipto Mangunkusumo General Hospital Jakarta from July 2016 to July 2017. We recruited ≥18 years old ME patients without a history of epilepsy or previous antiepileptic drug usage. Patients with focal neurological deficits were excluded. ME was defined as decreased consciousness associated with metabolic abnormalities such as decreased liver function (increasing liver enzymes more than five times normal range), renal function (serum creatinine level >10 mg/dL or blood urea nitrogen level >100 mg/dL), hypoxic-ischemic condition (e.g. acute global cerebral hypoperfusion; patient who had cardiac resuscitation and return of spontaneous circulation), electrolyte imbalance (e.g., serum Na level <115 mg/dL, serum Ca level < 5 mg/dL, serum mg level <0.8 mg/dL) blood glucose abnormality (serum glucose level <36 mg/dL or >450 mg/dL), and septicaemia (supported with hyper/hypothermia, biochemistry parameters e.g., leucocytosis [white blood cell count >12,000/μL] or leukopenia [white blood cell count <4,000/μL]).,, Informed consent was obtained from the patient's family.
EEG recording was performed to all patients that had been recruited in the emergency room, intensive care unit, and inpatient ward using Caldwell® ambulatory EEG machine with Arc® software. EEG was carried out for approximately 45–60 min to obtain at least 30 min of artifacts-free recording using the 10–20 International System. During the test, the level of consciousness was assessed by asking patients to do five things, which were: (1) “Say your name,” (2) “Repeat 1, 2, 3, ” (3) “Raise your arms,” (4) “Open your eyes,” and (5) “Look at examiner.” If no response was received from the subject, the procedure was repeated after giving strong tactile stimulation. NCSE was assessed based on Salzburg Criteria for Non-Convulsive Status Epilepticus (SCNC). The EEG recordings were reviewed and interpreted by two neurologists and certified electroencephalography readers.
According to SCNC, NCSE was classified as definite NCSE and possible NCSE. Definite NCSE was assessed if: (1) epileptic discharges (EDs) >2.5 Hz were seen in “worst” 10-second epoch, or (2) EDs ≤2.5 Hz or rhythmical delta theta (RDT) activity (>0.5 Hz) with presentation of at least one of following secondary criterion: (a) subtle clinical phenomena, (2) clinical and EEG improvement to intravenous epileptic drugs, (3) typical spatiotemporal evolution of the defining grapho-elements. Possible NCSE was assessed if: (1) there was fluctuation of grapho-elements without evolution, or (2) only EEG improvement of EDs or RDT activity with intravenous epileptic drugs was presented without clinical improvement.
All patients were observed subsequently after EEG examinations until the 30th day of admission. Patient outcome was classified as 'deceased' or 'survived'. Data analysis was performed using SPSS version 20. The statistically significant association was determined if P < 0.05 with 95% confidence interval. Ethical approval was obtained from the Ethics Committee of Universitas Indonesia – Cipto Mangunkusumo General Hospital.
Fifty patients with ME met the inclusion criteria and underwent the EEG examination. There were 32 (64%) patients diagnosed with NCSE based on EEG using Salzburg criteria. The demographic, clinical characteristics and outcome of all subjects can be seen in [Table 1].
Three common aetiologies of the ME in this study were sepsis, hyper-uremic, and hypoxia. Twenty-five patients had convulsive seizures before decreased consciousness (19 vs 6 in NCSE and non-NCSE group). Multiple aetiologies and convulsive seizures were more common in the NCSE group.
The EEG characteristics
In this study, all patients had slowing background rhythm (28 with theta wave background, 4 with delta wave background). Definite and possible NCSE diagnosis was determined by the SCNC criteria. The details of the EEG pattern were shown in [Table 2]. Only one of 12 subjects in definite NCSE had rhythmic epileptic discharge (ED) >2.5 Hz [Figure 1]. Two subjects showed ED <2.5 Hz and one of them had subtle ictal manifestation. Subtle ictal clinical manifestations were found in three subjects. All patients with possible NCSE showed only EEG improvement without clinical change during anticonvulsant administration. EEG improvement was defined as the reduction to “occasional”, i.e., 1–9% of epoch within 10 min after anticonvulsant fully applied.5 Periodic discharges and triphasic waves were seen in NCSE patients [Table 3]. Most of the NCSE patients had rhythmic delta/theta (RDT) activity.
There were only 21 patients of the NCSE (65.6%) group who were treated with antiepileptic drugs (AED) in maintenance dose until the 30th of admission. The rest of them were withdrawn from AED treatment due to some conditions (i.e., gastrointestinal bleeding) or side-effects (i.e., thrombocytopenia, anaemia). In Indonesia, the available intravenous AEDs are very limited (i.e., phenytoin, phenobarbital, diazepam, midazolam, propofol). Levetiracetam and valproic acid were administered orally through nasogastric tube.
The mortality rate in the NCSE and non-NCSE group was 40.6% vs 44.4%. All three subjects with subtle ictal manifestation were deceased, with one subject had ED <2.5 Hz and triphasic waves [Figure 2]. Most patients with triphasic waves were deceased and patients with periodic discharge were survived [Table 3]. One patient with the attenuated background was deceased.
Among the NCSE group, subjects who had single etiology were more likely to be survived compared to those with multiple aetiologies [Table 4]. Twenty-one NCSE patients received AED until day-30 and 13 of them (62%) were survived until the 30th day of admission.
Using the Salzburg Criteria for Nonconvulsive Status Epilepticus, the incidence of NCSE in ME in our study was 64%. About 60% of those patients had convulsive seizures before the NCSE diagnosis. Patients with ME who presented with convulsive seizures and had a decreased consciousness afterward should be suspected of experiencing NCSE. Differential diagnosis between ongoing ictal NCSE and postictal confusion was quite difficult to be made, especially in comatose ME patients who did not have a clear distinction between the non-convulsive ictal and postictal states. EEG monitoring played a significant role to determine the ongoing ictal activity in comatose patients.
All subjects who underwent EEG examination in this study were ME patients who remain unconscious despite resolved metabolic impairments. EEG patterns in ME were commonly found in form of triphasic waves (TWs), periodic discharges, and rhythmic delta activities (RDT). As shown in this study, TWs and RDT were commonly found and improved as anticonvulsant being administered. The next question would be whether the improved TWs with anticonvulsants met the diagnostic criteria of NCSE. A retrospective study reviewing 87 EEGs had found that TWs in encephalopathy were not improved with AED and yet, increased with stimulation whereas atypical TWs due to NCSE were improved and not altered by stimulation. Some authors considered NCSE in any patients with altered consciousness or confusional state with presenting atypical TWs. Atypical TWs were defined as “localized or lateralized sharp waves with triphasic configuration” and considered to be epileptogenic if they disappeared following AED treatment. Therefore, new terminology of TWs as “continuous 2/s GPDs with triphasic morphology” was used, especially in critical illness patients.
In this study, all of the subjects had slowing background rhythms. During metabolic abnormalities, there would be disturbances in the ascending reticular system and the afferent input to the thalamus of which is closely related to the alpha wave originating from the reticular nuclei of the thalamus. This problem is suggested to be the cause of background rhythm slowing as seen frequently in ME. Background slowing was widely present in various metabolic and endocrine abnormalities that presented with poor reactivity to photic, voice, and pain stimulation.
Despite previous studies about NCSE in altered mental state and consciousness patients, the study of NCSE which focuses on ME has not been available yet. Incidence of NCSE among altered conscious patients was widely varied between 8-37%., However, those studies did not refer to any specific etiology. These widely varied numbers were considered due to many EEG criteria used to identify NCSE.
The diagnosis of NCSE was often delayed because of its unrecognized subtle clinical sign. Symptoms of suspected NCSE could be identified by fluctuating mental status and subtle clinical signs such as lip-smacking, and nystagmoid eyes. NCSE was often initially misdiagnosed as a psychiatric disorder in patients presenting with altered mental status. NCSE could develop as de novo NCSE or occurring after a convulsive status epilepticus. In this study, about 40% of patients that were diagnosed with NCSE have not had any episodes of convulsive seizure before NCSE. Twenty-one out of 32 subjects in the NCSE group had GCS >8. Therefore, we had an important consideration that GCS might not be a reliable predictor for NCSE and EEG was critically important to establish a diagnosis of NCSE.
Various metabolic abnormalities often result in the disturbed function of the sodium-potassium channel in neurons which leads to the formation of epileptic discharge. This might be presented as convulsive or non-convulsive seizures in which both could be developed into status epilepticus. In this study, more than 50% of our patients with ME experienced convulsive seizures before unconsciousness. This number was considered similar to previous studies that reported about 30-40% of patients with convulsive seizures developed into NCSE with various aetiologies., This type of NCSE was suggested to increase the mortality and morbidity despite the etiology, although this result was still debated., Distinguishing ongoing NCSE from the clinical postictal state can be difficult in some cases. Prolonged confusion is common following tonic-clonic seizures. This symptom along with ictal discharges in EEG recording was considered as NCSE. In comatose patients, a clear distinction between the ictal and postictal states is difficult and impossible to differ without EEG examination. Subtle motor activities are the clinical indicators of ongoing ictal activity, which have to be promptly managed with antiepileptic agents.
In this study, we found 3 subjects presented with subtle ictal manifestation which had a repetitive epileptiform discharge <2.5 Hz in EEG examination in 1 patient and RDT >0.5 Hz in 2 patients. All subjects with subtle ictal manifestation were deceased. Subtle ictal manifestations were often overlooked. This symptom indicated an ongoing electrical ictal activity and may lead to refractory NCSE. At the cerebral tissue level, this symptom demonstrated a maladaptive and ongoing hyper-excitable process leading to neuronal injury.,
Several ME case reports presenting NCSE had been reported in hepatic, uremic, sepsis and hypoxic encephalopathy.,, A retrospective study of septic patients in the medical intensive care unit undergoing EEG monitoring found the electrographic seizures in 16% and periodic epileptiform discharges in another 16% patients. In our study, almost 70% of the NCSE group were suffered from multiple aetiologies with sepsis as the most common cause.
The mortality rate of NCSE and the non-NCSE group was similar in our study. Despite EEG improvement, the clinical improvement could not be achieved in NCSE subjects after AED administration. It explained that treating the underlying etiology was also important to decrease the mortality rate. Aggressive anti-epileptic treatment should be carefully considered since AED may contribute to morbidity and mortality of critically ill patients. Encephalopathy in sepsis patients is associated with a higher mortality rate, a higher rate of electrographic seizures and poorer outcome. Continuous EEG (cEEG) was associated with increased initial seizures detection and modification of AED but was not related to improved outcome. Level of consciousness could not predict the mortality among NCSE subjects in our study, differed with ME patients in another study.
Neither cEEG nor repetitive EEG was performed to detect ongoing electrographic seizures in this study due to our source limitation. Therefore, follow-up and treatment evaluation were only based on clinical assessment and not by the detection of electrographic seizures.
Incidence of NCSE among patients with ME at our hospital was high. Despite the anti-epileptic treatment of the NCSE group, the underlying cause of ME was still the main factor that influences the outcome. Therefore, aggressive treatment of AED should be very carefully considered knowing the possible side-effect that might worsen the outcome of patients with ME.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
All named authors have met the International Committee of Medical Journal Editors (ICMJE) criteria for authorship, took responsibility for the integrity of the work as a whole and have given their approval for this version to be published.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]