Received Date: October 14, 2015; Accepted Date: May 31, 2017; Published Date: June 07, 2017
Citation: Liu TT, Wang J, Lu GP, Wang Y (2017) Analysis of 247 Children with Status Epilepticus: Clinical Features and Prognosis. Emerg Med (Los Angel) 7: 354. doi:10.4172/2165-7548.1000354
Copyright: © 2017 Liu T, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License; which permits unrestricted use; distribution; and reproduction in any medium; provided the original author and source are credited.
Visit for more related articles at Emergency Medicine: Open Access
Background: Status epilepticus (SE) is one of the most common neurological emergencies in childhood. However, the clinical features and prognosis are poorly characterized in China. Objectives: To summarize the clinical features and outcomes among Chinese children. Methods: Children aged from 29 days to 18 years with SE were included in Children’s Hospital of Fudan University. The demography, etiology, seizure types, treatment and outcome were analysed. The statistics were conducted by SPSS19. Results: There were 247 SE patients (F:M=1.25:1). The mean age was 2.4 years. Most of them had acute symptomatic SE (53.06%) including viral encephalitis (48.46%). Most of them had convulsive status epilepticus (83.4%), others had non-convulsive status epilepticus (16.6%). Most of them had combination therapy (74.89%). Lengths of follow-up ranged from 1 month to 3 years after onset of SE. Most of cases were bad outcome (73.80%). Etiology and neurological image were significant risk predictors by multiple logistic regression analysis. Mortality rate was 16.95%. Recurrence rate was 20.76%. Most of cases died within 1 month after onset of SE (65.0%). Most of cases also recurred within 1 month (81.63%). Duration of SE and mechanical ventilation were related to death by multiple logistic regression analysis. The number of antiepileptic drugs (AED) and anaesthetic treatment were related to recurrence of SE by Cox survival analysis. Conclusion: The proportion of SE was highest in children younger less than 1 year. The most common etiology of SE in children was acute symptomatic, especially viral encephalitis. The clinical features between CSE and NCSE were different. Risk factors between recurrence and death were different. The death rate was higher in cases with duration of SE ≥ 3 hours and mechanical ventilation. Maintenance treatment ≥ 3 AEDs and anaesthetic treatment were related to higher recurrence rate. These results suggested we can improve the prognosis by terminating seizure in time
Status epilepticus; Clinical features; Outcome
Status epilepticus (SE) is one of common medical emergencies of the paediatric age group. It will lead to serious and even irreversible neurological abnormalities including death if not diagnosis and treatment properly. The traditional epidemiologic definition of SE as continuous or intermittent seizures lasting for more than 30 min with incomplete recovery of consciousness may be impractical . However, more recent definitions advocate a much shorter operational definition: either continuous seizures lasting at least 5 min or two or more discrete seizures between which there is incomplete recovery of consciousness [2,3]. SE has been classified based on clinical features, such as convulsive SE (CSE) or non-convulsive SE (NCSE).
The incidence of CSE in children is approximately 10–27/100,000 per year, and is higher than the 4-6/100,000/year reported in epidemiologic studies of convulsive status epilepticus [4,5]. NCSE have provided overall population incidence rate of between 5.6 and 18.3 per 100000 individuals per year. NCSE often follow CSE, accounting for 26% of the NCSE. The present study focuses on adults, but its research data may not be able to provide children SE reliable basis for clinical work. The children as a special group, the incidence, mortality, etiology, SE type, treatment and prognosis are different from adults . To recognize etiology, seizure type, treatment, rise factors, outcome and the correlation between each other of children with SE is significant for management. Unfortunately, there are few published data from population-based and hospital-based studies of paediatric groups in China.
This is an observational study. In this study, we examined patients 29 days to 18 years who were diagnosed with SE in our hospital between 1 January 2012 and 31 December 2014. SE is defined as continuous or intermittent seizures lasting for more than 30 min without complete recovery of consciousness, including CSE and NCSE. Exclusion criteria: a seizure is less than 30 minutes; the new-born and more than 18 years old.
We queried cases with a diagnostic code of SE via usage of the International Classification of Diseases, 10th revision (ICD-10) code G41.900. We collected detailed information from medical record, including demography, etiology, seizure types, EEG, VEEG, MRI/CT and treatment. Lengths of follow-up ranged from 1 month to 3 years after onset of SE by electronic medical record (EMR) and telephone. The outcome include neurological sequelae, recurrence and death following SE.
The seizure types were classified as CSE and NCSE. The etiology types are listed and types in Table 1 [6,7]. Outcomes of SE patients and related factors were evaluated and described using the Glasgow Outcome Score (GOS), including good (GOS=1-4) and bad (GOS=5).
|Febrile SE||SE occurring when the only provocation is a febrile(38.0?C) illness and in the absence of defined CNS infection.|
|Acute-symptomatic||SE in a previously neurologically normal child, within a week of an identified acute neurological insult.|
|Remote-symptomatic||SE in a child with a history of a preexisting
CNS abnormality in the absence of an identified acuteinsult.
|Acute-on-remote-symptomatic||SE occurring during an acuteneurological insult or febrile illness in a child with a remotesymptomatic etiology.|
|Cryptogenic/idiopathic epilepsy-related||SE that occurred in children with a previous diagnosis with idiopathic/cryptogenic epilepsy, or SE that is the child’s second unprovoked seizure.|
|Unclassified||SE that cannot be classified into any other group.|
Table 1: The etiology types of SE.
SPSS19.0 software was used for statistical analysis with p-values of <0.05 and was considered significant. A chi-square test was conducted. We used logistic regression and Cox survival analysis to analyse the clinical outcome.
In this study, 247 children were included. Details are shown in Table 2.
|Item||Inpatientn (%)||Outpatientn (%)||Total n (%)|
|First ever episodes|
|Cerebrospinal fluid (CSF)|
|One kind of AEDs||44(20.28)||3(30.00)||47(20.70)|
|Two kind of AEDs||55(25.35)||3(30.00)||58(25.55)|
|= Three kind of AEDs||109(50.23)||3(30.00)||112(49.34)|
|One kind of AEDs||72(30.51)||-||72(30.51)|
|Two kind of AEDs||47(19.92)||-||47(19.92)|
|= Three kind of AEDs||76(32.20)||-||76(32.20)|
|Mild hypothermia therapy||17(7.20)||-||17(7.20)|
Table 2: Detailed information of SE.
There were 247 SE patients including in this study, including 236 inpatients and 11 outpatients. 132 of 236 inpatients came to paediatric intensive care unit (PICU). The ratio of males to females was 1.25:1. The mean age was 2.4 years, with 58.30% less than 3 years of age.
245 of 247 subjects can clear the etiology. Within our patients, most of them had acute symptomatic SE (130, 53.06%) including viral encephalitis (48.46%), followed by acute-on-remote-symptomatic SE in 54(22.04%) cases. 28(11.43%) cases were remote symptomatic SE 28(11.43%). Febrile SE accounted for 16(6.12%) cases. Others were unclassified cases (3, 1.22%).
Within our patients, most of them had CSE (206/247, 83.4%). Others had NCSE (39/247, 16.6%). Most of CSE were generalized seizure (144/202, 71.3%) including tonic-clonic seizure (65.3%), tonic seizure (31.2%), clonic seizure (2.8%), myoclonic seizure (0.7%). 56.2% CSE were intermittent seizure. 47.3% CSE lasted 30 min-1 h. Most of NCSE were partial seizure (73.2%). 85.4% NCSE were intermittent seizure. 85.4% NCSE lasted longer than 3 h. Most of NCSE were diagnosed by VEEG (95.12%).
In this study, abnormal cranial imaging results mainly include signal changes (such as cortex, thalamus, basal), brain dysplasia, intracranial haemorrhage, vascular disease (vascular malformations, cerebral infarction), intracranial placeholder, cerebral edema and hypoxia ischemic injury, etc. Statistical analysis showed that cerebral dysgenesis (χ2=15.801, P=0.003), cerebral edema (χ2=15.520, P=0.002), ventricular expansion (χ2=8.276, P=0.036) and brain atrophy (χ2=10.780, P=0.012) with bad prognosis.
184 of 227 EEG cases had video electroencephalogram (VEEG). Statistical analysis showed that bad outcome were related to frequent discharge (χ2=17.87, P=0.001), the PLEDs (χ2=8.60, P=0.037), three phase wave (χ2=9.913, P=0.005).
AEDs include clonazepam (vein), phenobarbital (intramuscular injection, intravenous and nasogastric), chloral hydrate (oral and enema), midazolam (vein), diazepam (vein), valproate (intravenous and oral), levetiracetam (oral), topiramate (oral), etc. The most of them was clonazepam (209 cases).
Within our patients, 4.41% without AEDs can relieve themselves for failing to see a doctor. Most of them had combination therapy (74.89%). 40.0% (99cases) were refractory status epilepticus (RSE). All of RSE were combination therapy. 14.14% RSE used anaesthetics. 34.75% inpatients used mechanical ventilation. The mean ventilation time was 8 days.
Hospital charge and duration of hospitalization
Mean ± SD duration of hospitalization was 17.81 ± 17.35 days (range, 1–114 days), and median (P25, P75) direct hospital charge was equal to 20165.51(9178.33, 43325.81) RMB. Direct hospital charge was significantly longer for patients with RSE than for those with SE (P=0.004). Duration of hospitalization was significantly shorter for patients with good outcome (GOS=5) than for those with bad outcome (GOS=1-4) (P=0.005).
Lengths of follow-up ranged from 1 month to 3 years after onset of SE. Most of cases were bad outcome (73.80%). During the follow-up, outpatients were excepted. Data after onset of SE are shown in Table 3.
|Cranial imaging (n=48)|
|Intellectual assessment (n=23)|
|One kind of AEDs||49||30.82%|
|Rehabilitation therapy (n=187)||41||21.93%|
Table 3: Data after onset of SE.
Single factor Logistic analysis of P<0.05 items include CNS diseases history, etiology, SE types, SE forms (partial/generalized seizures), duration of SE, neurological image, the kind of AEDs, recurrence. With multiple logistic regression analysis, causes, neurological image were selected as independence predictors of SE risk (Table 4).
|Cryptogenic/idiopathic epilepsy-related||2.229||0.570, 63.603||6.023||0.135|
|Abnormal neurological image||3.299||1.157, 9.405||4.988||0.026|
Table 4: Comparison of time intervals.
Mortality rate was 16.95% (40/236). Most of cases died within 1 month after onset of SE (65.0%). According to single-factor logistic regression analysis, CNS disease history, seizure character (continuous/ intermittent), duration of SE, combined anesthesia, mechanical ventilation, recurrence were related to Death. To further analysis, the death rate was higher in cases with duration of SE ≥ 3 hours and mechanical ventilation by multiple logistic regression analysis (Table 5).
|CNS disease history||0.440||0.140, 1.381||1.980||0.159|
|Continuous seizure||1.487||0.418, 5.291||0.375||0.540|
|Duration of SE|
|Combined anesthesia||1.795||0.328, 9.817||0.455||0.500|
|Mechanical ventilation||4.385||1.696, 11.200||9.349||0.002|
Table 5: Multiple logistic regression analysis for SE.
Recurrence rate was 20.76% (49/236). Most of cases also recurred within 1 month (81.63%). According to single-factor Cox survival analysis, SE types, duration of SE, seizure character (continuous/ intermittent), the kind of AEDs, combined anesthesia and mechanical ventilation were related to recurrence. Finally, sequential therapy ≥ 3 AEDs and anesthetic treatment were related to higher recurrence rate by multiple Cox survival analysis (Table 6). Kaplan-Meier curves of recurrence showed in Figure 1.
|Seizure features(mode 1)|
|Duration of SE|
|Continuous seizure||0.609||0.265, 1.402||1.385||0.244|
|Single AEDs||0.434||0.084, 2.238||0.993||0.319|
|Two kind of AEDs||1.541||0.372, 6.377||0.356||0.551|
|=three kind of AEDs||3.462||0.994, 12.060||3.802||0.051|
|Combined anesthesia||3.722||1.512, 9.160||8.178||0.004|
|mechanical ventilation||1.648||0.760, 3.570||1.601||0.206|
|Duration of SE|
|Single AEDs||0.508||0.102, 2.535||0.682||0.409|
|Two kind of AEDs||2.001||0.491, 8.159||0.935||0.334|
|=three kind of AEDs||3.760||1.066, 13.256||4.243||0.039|
|Combined anesthesia||3.600||1.632, 7.942||10.074||0.002|
Table 6: Multiple cox survival analysis for recurrence.
Status epilepticus is the most common childhood medical neurological emergency. Most data for this disorder are from mainly adult populations and might not be relevant to childhood. At present, domestic related research is incomplete. We required data from Children’s Hospital of Fudan University. We hope it will provide evidences for the current clinical practice and further research.
Within our patients, the ratio of males to females was 1.25:1. This showed that bays ratio is higher than girls. De Lorenzo et al.  reported the same results. However, the conclusion is debatable. The most patients were aged younger 1years following cases aged from 1 to 3years. Chin et al.  showed the incidence of SE was highest in children younger than 1 year. Nishiyama et al.  developed multicenter epidemiologic study. The results showed that the highest incidence was seen in patients aged <2 years, especially in the second year of life. Lin et al.  reported the highest incidence of SE was in the 1-5 year age group. Approximately two thirds of the patients were less than 5 years old when they manifested SE.
In this study, most of them had acute symptomatic SE including viral encephalitis, followed by acute-on-remote-symptomatic SE. Febrile SE accounted for only 6.12%. A number of studies show that febrile SE is the most common cause in developed country [6,10]. However, a study in India showed their conclusion is the same with us. The causes of children SE may be different. Between developed countries and developing countries. In addition to, the differences may be the reason for this study is given priority to with hospitalized children, and most of children were from PICU.
Within our patients, most of them had CSE (83.4%). Others had NCSE (16.6%). Most of NCSE were diagnosed by VEEG. The clinical manifestations of NCSE in children typically include cognitive or behavioral impairments. The availability of VEEG monitoring has allowed for increased recognition of this condition [11-13]. Seizures in the acute setting, ocular movements and ongoing CNS infection are associated with NCSE . In China, children NCSE Still need further study.
With multiple logistic regression analysis, causes, neurological image were selected as independence predictors of SE risk. Asadi- Pooya et al.  reported outcome was significantly correlated with etiology of SE. This result is same with us. Lin et al.  reported that only the etiology and age were significant predictors of outcomes. SE with overall prognosis is poor , but significant predictors vary in different research. In a word, it remains to be further discussed.
Death of SE
In our study, mortality rate was 16.95%. The death rate was higher in cases with duration of SE≥3 hours and mechanical ventilation by multiple logistic regression analysis. Some studies showed the significant predictors included age, etiology, duration of SE and effective treatment [18-21]. Within our study, most of cases died within 1 month after onset of SE. Ostrowsky et al.  showed short-term (during the first 30-60 days after SE) mortality rates of 3% to 5%. It was different from our conclusion. Domestic related data is rare. And it needs to be further discussed.
Recurrence of SE
Recurrence rate was 20.76%. Sequential therapy ≥3 AEDs and anesthetic treatment were related to higher recurrence rate by multiple Cox survival analysis. Multivariate analyses showed that progressive encephalopathy and preexisting neurologic deficit were independent predictors for SE recurrence . Pediatric related data is rare. An adult study showed recurrence risk is independently associated with chronic etiology and to a lesser extent with female gender . Followup time, etiology classification, object of study, resulted in different research results. It is necessary to further study.
The proportion of SE was highest in children younger less than 1 year. The most common etiology of SE in children was acute symptomatic, especially viral encephalitis. Duration of SE, continuous and intermittent seizure were related to the etiology.
The clinical features between CSE and NCSE were different. The characteristics of NCSE were variable. We can use VEEG to improve the rate of diagnosis.
Risk factors between recurrence and death were different. The death rate was higher in cases with duration of SE ≥3 hours and mechanical ventilation. Maintenance treatment ≥ 3 AEDs and anesthetic treatment were related to higher recurrence rate. These results suggested we can improve the prognosis by terminating seizure in time.