- Top of page
Summary: Background: The incidence of status epilepticus (SE) in Asian children, including Japanese, has not been reported.
Methods: In 2003, we performed an epidemiological study of SE on Japanese children (31 days or older to <15 years of age) in Okayama City by ascertaining all lifetime first episodes of SE.
Results: Thirty-seven patients (22 males and 15 females) were identified. The annual incidence of SE was 38.8 per 100,000 population (95% CI: 24.5–49.5). Febrile SE in the absence of CNS infection accounted for 17. Acute symptomatic etiologies other than febrile SE were observed in eight patients, including three cases of influenza encephalitis/encephalopathy. Five were classified as remote symptomatic and the remaining seven as cryptogenic. The highest incidence (155.1/100,000) was seen in the age range of 31 days or older to <1 year, followed by 101.5/100,000 in the age range of one year, and the incidence decreased after eight years. In 26 of the 37 patients, SE was their first seizure. As for seizure types, 32 had convulsive SE, including tonic status in one. Five others showed nonconvulsive SE, including complex partial SE in four and absence status in one. No one died of SE. Two patients who brought on SE because of influenza encephalitis/encephalopathy suffered from motor disturbance with or without mental disturbance after SE.
Conclusions: The incidence of SE tended to be higher in Japanese children than reported in Caucasians. The Japanese had an age-specific incidence pattern similar to that of Caucasians.
Status epilepticus (SE) is among the more common pediatric neurological emergencies requiring immediate aggressive intervention. To establish SE's best management, we must first clarify its incidence and characteristics in the general population. A previous epidemiological study (DeLorenzo et al., 1996) in the United States suggests the presence of racial differences in the incidence of SE; however, no such studies have been conducted in Asian countries, including Japan. To determine the annual incidence of the first occurrence of SE in Japanese children and to elucidate its characteristics, we performed an epidemiological study on it in Okayama City in 2003.
Okayama City is an appropriate place for surveys of this kind, since we have already performed several population-based neuroepidemiological surveys in Okayama Prefecture (Ishida, 1985; Oka et al., 1995; Ohtahara et al., 1997; Oka et al., 2006). The city is the central city in that prefecture and has ten general hospitals, including Okayama University Hospital. We have a 24-h emergency medical care system for children in cooperation with these hospitals. The area of this city is 513.29 km2, and it has no high mountains or islands.
- Top of page
This is a retrospective study. Included were all patients aged 31 days or older to <15 years of age living in Okayama City who had developed a lifetime first SE from January 1 to December 31, 2003. The subjects were recruited from all hospitals in the above-mentioned emergency system to which patients with SE might visit. We collected detailed information on them that included age, gender, past history of SE and seizures, etiology of SE, developmental history, underlying diseases, family history of seizures, duration of SE, seizure types, laboratory data, EEG findings during SE and/or after SE, cranial CT and/or MRI findings after SE, medical treatment in the acute period of SE, sequelae after SE, and the degree of sequelae, if any. These anonymized data were obtained from doctors in charge of the patients by means of a detailed questionnaire in January 2004. Just after receiving the questionnaire responses, we thoroughly examined all data and confirmed that there were no case duplications. We discussed all uncertainties with the doctors in charge of the patients, and one of the authors (I.N.) reviewed medical records to evaluate the preciseness of the data. This procedure had been completed by the end of June 2004. Most subjects came to the hospital as emergency cases and were admitted after arrival for treatment and thorough examinations, such as EEGs and neuroimaging studies. Besides these patients, we included inpatients who had had SE during hospitalization in children's wards and those in other wards, such as intensive care units. We tried to collect data through pediatricians and pediatric neurologists who were consulted for the treatment of SE of children in wards other than children's wards.
The design of this study was approved by the Ethics Committee of Epidemiology in the Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences.
SE was strictly defined as any seizure lasting for 30 min or more or intermittent seizures lasting for more than 30 min from which the patient did not regain consciousness. Nonconvulsive SE was also included (Table 1). With respect to etiologies, the acute group consisted of those with acute symptomatic etiologies and febrile SE (≥38°C). The nonacute group included those with remote symptomatic etiologies and cryptogenic cases.
Table 1. Classifications of seizure types
|I Convulsive SE|
|(1) Generalized convulsive SE: generalized tonic, clonic, and tonic–clonic seizures|
|(2) Partial convulsive SE: convulsions with focal features, including secondarily generalized seizures|
|II Nonconvulsive SE: seizures in which the main symptom is the disturbance of consciousness with or without subtle motor manifestations, such as eye deviation lasting for 30 min or more|
|(1) Complex partial status|
|(2) Absence status|
The definition of seizure types is described in Table 1. The differential diagnosis between complex partial status and absence status was made by the clinical observation and interictal EEG findings and, whenever possible, by ictal EEG findings.
Age-specific incidence was calculated, dividing the number of cases by the age-specific number of the population. A 95% confidence interval (CI) was estimated, assuming Poisson distribution. Fisher's exact test was used to compare the difference between the acute and the nonacute groups. We used the statistical significance level of p = 0.05.
- Top of page
In this study, the annual incidence of SE in children aged 31 days or older to <15 years was estimated to be 38.8 patients per 100,000 population (95% CI: 24.5–49.5). In Japan, the total number of children aged <15 years is about 18,000,000. Based on this estimation, it is presumed that at least 7,000 new cases of SE per year appear in this age group in Japan.
This is the first epidemiological study on SE ever reported by an Asian country. In Western countries, four such studies are available (DeLorenzo et al., 1996; Hesdorffer et al., 1998; Coeytaux et al., 2000; Chin et al., 2006). Three of them in a predominantly Caucasian population have provided to be very consistent estimates of the incidence of SE in children (<15 or <16 years of age). These studies report an incidence figure of 17–24.1/100,000. The remaining study performed in a racially mixed population could provide information on Caucasians separately from other racial groups. Notably, this study reported a substantially higher rate among non-Caucasians (57/100,000), compared with 20/100,000 for Caucasians in all age ranges (DeLorenzo et al., 1996). In our study, the incidence of SE for Japanese children tended to be higher than in Caucasian children, but it was lower than in non-Caucasian children of the United States (DeLorenzo et al., 1996).
Several other factors besides race might also contribute to the high incidence of SE. In Japan, we have a social health insurance system that basically covers everyone. All medical fees for children aged <4 years are paid by the Okayama City government, and all residents have easy geographical access to general hospitals in the emergency medical system. The ambulance system is well established, and all residents can use it for free; thus all children in Okayama City have easy economic and geographic access to high-level medical service. Accordingly, parents are able to take children with seizures to the general hospitals whenever necessary, and an accurate diagnosis of SE can be made by pediatricians or pediatric neurologists at any time. We suspect that might be one reason why we are able to identify many patients with SE. Japan is one of the world's most industrialized countries and it is quite unlikely that cultural, dietary, or religious factors can contribute to a high incidence of SE.
As for etiology, the percentage of febrile SE is high in the worldwide pediatric population. In Western countries, it is thought that febrile SE accounts for approximately one quarter of all childhood SE (Shinnar and Glauser, 2002). In our study, the percentage of febrile SE reached 46%. The prevalence rate of febrile seizures in Japan (Tsuboi, 1984) is much higher than in the United States (Nelson and Ellenberg, 1978) and other countries (Verity et al., 1985; Doerfer and Wässer, 1987; Okan et al., 1995). Therefore, the high percentage of febrile SE in our study might be a result of racial difference of febrile seizures.
Another finding of interest in our investigation is that influenza-related SE was the biggest cause of acute SE. In Japan, febrile seizures and febrile SE resulting from influenza are frequently seen. Moreover, the number of children diagnosed as acute influenza encephalitis/encephalopathy exceeds 100 per year (Morishima, 2003). The incidence of influenza encephalitis/encephalopathy is higher in East Asia than in Western countries. This may reflect differences in the epidemiology of influenza or possibly racial differences in susceptibility to its effects.
In the nonacute group of our study, many patients suffered from epilepsy. SE was the initial unprovoked seizure in 6 of the 12 in this group, including one with a history of four febrile seizures before the SE. Berg et al. (1999) reported that 56 (9.1%) of 613 children with epilepsy had had one or more episodes of SE by the time the diagnosis of epilepsy was established. Most SE episodes had occurred as the first or second unprovoked seizure. Sillanpää and Shinnar (2002) also found that the risk of SE was highest at the onset of the epilepsy in a population-based cohort with childhood-onset epilepsy in Finland. These facts indicate the difficulties in predicting SE in children with epilepsy as well as in those with acute symptomatic etiology and febrile SE.
Comparing the results of our study with the previous Western reports on SE (Aicardi and Chevrie, 1970; Maytal et al., 1989; DeLorenzo et al., 1996; Shinnar et al., 1997; Hesdorffer et al., 1998; Coeytaux et al., 2000; Chin et al., 2006), we found several similarities in the characteristics of childhood SE patients. The annual incidence in children aged <2 years of age, especially during the first year of life, was the highest throughout all age ranges. With regard to the mortality and morbidity of SE in children, the results of our study were compatible with others, indicating that the occurrence of either morbidity or mortality of SE in childhood was not high in the absence of an acute neurological insult. A large proportion of patients with benign symptomatic etiologies and febrile SE and the easy access to high-level medical service might explain why the number of severe cases with an unfavorable outcome was small in spite of the high-incidence rate of SE in our study.
This is the first population-based epidemiological study reported not only from Japan but also from Asian countries. The results indicate a tendency for the incidence of SE to be higher in Japanese children than in those reported in Caucasians, and we suspect that the high incidence might be partly related to racial difference. This study has several limitations. One is that it is a retrospective study. Another might be a slight possibility of underestimating the subjects, since we collected data mainly through pediatricians and pediatric neurologists. Another limitation is that we investigated the occurrence of SE for only one year. The number of cases identified was thus relatively small, and a 95% CI became large in the statistical analyses. Large-scale epidemiological studies are needed to confirm our conclusion.