1. Background

Status epilepticus (SE) is the most extreme form of seizure that may occur in patients with previous epilepsy or acute disorders of the central nervous system. SE is a medical emergency that may lead to permanent brain damage or death. SE is very rare, and most people with epilepsy will never experience this disorder [1–4]. The estimated annual incidence of SE in children ranges from 10 to 38 per 100,000, and its prevalence is higher in younger children, especially in the first year of life. The estimated mortality in SE patients is 20%. SE can occur as convulsive SE or nonconvulsive SE. Many factors determine the prognosis of this disorder; among these major factors are the duration of SE, patient age, and the underlying cause. Common causes of SE include cerebrovascular disorders, brain trauma, infections, low antiepileptic drug levels in patients with epilepsy, inflammatory causes, inborn errors of metabolism, and genetic causes. In this study, we aimed to determine the prognostic factors for SE in our patients and to test the availability of the modified status epilepticus severity score (mSTESS) system [3–10].

2. Methods

The medical records of patients aged 1 month–17 years who were treated at Çukurova University–Balcalı Training and Research Hospital from September 2018 to September 2021 were retrospectively reviewed to detect the following ICD-10 codes: G41.0, grand mal SE; G41.1, petit mal SE; G41.2, complex partial SE; G41.8, other SE; and G41.9, SE unspecified. Patients with the abovementioned ICD codes were included in the study. Demographic and clinical variables, including age, sex, age of onset of SE, type of seizure, etiology, available genetic analysis results, EEG results, magnetic resonance imaging (MRI) results, mSTESS, received treatments, responses to treatments, developmental status at baseline before SE and after SE (before discharge at the last available follow-up visit), and mortality, were recorded [11, 12]. SE was diagnosed according to the International League Against Epilepsy (ILAE) 2015 classification. All noncompliant seizures, psychogenic nonepileptic seizures, and the newborn population were excluded from the study. Patients with severe disabilities and children with acute hypoxic–ischemic encephalopathy were also excluded from the study. Approval from the institutional ethics committee was obtained. Written informed consent was obtained from the parents/caregivers.

3. Results

Seventy-two patients were included in this study. The mean age of the patients with SE was 5 years [3–8]. The male-to-female ratio was 34 (47%)/38 (53%). A history of epilepsy was present in 53% of the patients (Table 2). The most common etiologies underlying SE were meningoencephalitis (19%), febrile status (8%), unknown (12%), and genetic causes (7%). Tables 2, 3, and 4 show the characteristics of the patients, the etiology underlying SE (Table 2), and factors that played a role in the good and poor prognoses of SE (Table 4). Comorbidities, including developmental delay/intellectual disability, cerebral palsy, hyperactivity, and autism spectrum disorder, were present in 44 (61%) of the patients with SE. The case mortality rate was 1% in our study. ROC curves depicting the prognostic accuracy of mSTESS (in all of the patients) were generated (Figure 1). The classification accuracies of mSTESS according to different cutoff points and the predictive accuracy of the mSTESS score are also shown (Tables 5 and 6).

4. Discussion

SE is a common neurological emergency. The incidence of short-term mortality ranges from 0.9% to 3.6% in children. The most vulnerable structure to seizures is the hippocampus, which is involved in learning and memory. Other structures that demonstrate necrosis following SE include the amygdala, dorsomedial thalamic nucleus, medial layers of the neocortex, and cerebellum; thus, SE has a destructive effect on cognitive, motor, and sensory functions [1–4]. SE outcome is mainly determined by any delay in treatment, the refractory degree of the seizures, the underlying etiology, and the response to treatment. The determination of the factors that can predict the outcome of patients with SE is highly important because these factors may be useful and essential for deciding on any further treatment that can directly affect patient prognosis [12–15].

Age is a strong indicator of outcome in SE patients. Febrile seizures and acute symptomatic etiologies are most common in children younger than 2 years, whereas cryptogenic and remote symptomatic etiologies are more common in older children. According to some studies, mortality due to SE is high in children younger than 1 year, which is possibly due to the fact that the brain is in a period of rapid development at that age [16]. Our study demonstrated that age did not play a role in the prognosis of SE, which was likely due to the fact that most of our patients younger than 2 years had SE due to acute symptomatic etiologies such as meningoencephalitis, febrile SE, and an unknown origin. The effect of age could be due to the fact that the etiology of SE differs with age. Sex can also affect the prognosis of patients with SE. According to a study conducted on adult patients to assess the role of gender in SE, the hospital-based prevalence of SE appeared to be greater for men than women. The hospital stays of men with SE were shorter than those of women, which indicates that the severity of SE tends to be greater in females who require prolonged hospitalization [17]. In our study, the mean age group was 5 years, and sex did not play a role in the prognosis of SE patients. This may be because the estrogen and testosterone hormones do not affect young males or females differently as it does in males or females after puberty.

Ethnicity is also a factor that can play a major role in prognosis of this disorder. There is a higher incidence of SE and lower mortality among black individuals. This could be related to the underlying illness, access to medical care, compliance, or other intrinsic biological factors. In our study, all of the included patients were Asian; therefore, such a comparison could not be performed [18].

An impaired level of consciousness at SE onset is independently associated with mortality. In a recent retrospective population-based study of adult patients with SE in Salzburg, impaired consciousness was associated with a high case fatality rate of 33%, compared with 8.2% in awake patients [18, 19]. In our study, 47% of the patients (alert, confused, and somnolent) were in the good prognosis group, and 4% were in the poor prognosis group, whereas 53% of the patients (comatose and stuporous) were in the good prognosis group, and 96% were in the poor prognosis group. This scenario could be explained by the destruction of brain cells due to prolonged seizures. Semiology could also play a role in the prognosis of SE patients. According to a previous study, generalized myoclonic SE is associated with poor outcomes and high mortality. We have noted from the STESS that generalized seizures have a worse prognosis than focal seizures. Our study matches these results [18, 19].

It is difficult to determine the exact duration of SE because its onset is frequently not known. Overall, a longer duration of SE was associated with greater mortality. However, new-onset refractory SE can be associated with significant recovery even after a prolonged duration of SE. In our study, the duration of SE could not be obtained from the medical files; however, we followed a male patient aged 10 years with SE in the ICU due to febrile infection-related epilepsy syndrome (FIRES). He had very prolonged durations and frequent SE despite the fourth-line treatment protocol for SE. Furthermore, he had no seizure-free periods, and he died 9 months after admission to the ICU due to complications.

Overall, EEG findings were less important than the duration of SE and etiology as predictors of outcomes. The EEG pattern is nonspecific in children with SE. In a systematic review of adult patients with SE, periodic epileptiform discharges (PEDs) were found to be associated with poor outcomes. According to an EEG study of 50 patients, PEDs were associated with greater mortality/vegetative state (44%) than were non-PEDs (19%). PEDs mainly occur due to diffusion restriction in the cortical area, which is commonly observed in SE. Normalization of the EEG after SE was correlated with good outcomes [18, 19]. In our study, patients with slow-wave EEGs had a worse prognosis than patients with normal EEGs or with spikes and/or sharp waves.

Huang et al. studied 15 patients with SE via MRI. These MRI abnormalities consisted of decreased diffusion in diffusion-weighted imaging (DWI)/apparent diffusion coefficient (ADC), and increased signal in T2 MRI sequences was usually accompanied by focal cerebral edema and increased vascularization [20–25]. In our study, seven (9.7%) patients had increased T2 hyperintensity and decreased diffusion in the ADC. The failure of the Na/K ATPase pump may lead to cellular sodium and water influx, thus resulting in cytotoxic edema. Other mechanisms may include the excessive release of excitatory amino acids such as glutamate and increased membrane ion permeability [16–19]. In our study, it was concluded that the presence of abnormal MRI–CT results was associated with poor outcomes in children with SE.

Metabolic imbalance plays a crucial role in prognosis. A decrease in glucose and glycogen with parallel increases in lactate indicates a high rate of glycolysis, and seizures can result in profound elevations in serum lactate. In a study conducted by Sop et al. on adult patients with seizures regarding the relationship between high blood lactate levels, mortality, and poor prognosis, no relationship could be established. In our study, we concluded that patients with a poor prognosis mainly had a higher lactate level, yet the mean average blood lactate level was less than 2 mmol/l, which can be considered normal because in other studies, when high and low lactate levels were classified, a cutoff of 2 mmol/L was used as an indicator of a high lactate level. The difference between the results of these studies could be due to many factors, such as the time at which blood samples were obtained, as well as the fact that our study population involved a pediatric group and the presence of other comorbidities, such as infections, which could have affected blood lactate levels [26, 27].

Associated conditions can always complicate the course of SE and are considered important prognostic factors for cardiac arrhythmia, cardiac damage because of catecholamine surge, and respiratory failure. The presence of these complications can also affect the prognosis of patients with SE. In our study, there was no effect of low blood pressure, arrhythmia, or intubation on poor prognosis; however, in our study, there was a male patient aged 10 years followed by FIRES who experienced low blood pressure, arrhythmia, intubation, and multiple organ failure. He died 9 months after admission to the pediatric ICU. Moreover, he experienced super-refractory SE despite the fourth-line treatment protocol for SE, which included anakinra ketogenic diet and cannabidiol treatment [28, 29].

Comorbidities such as cerebrovascular disease, renal failure, hyperactivity, and autism can coexist with SE in children. These comorbidities can affect the prognosis of patients with SE. Most of the studies researching the effect of comorbidities on the prognosis of SE have been conducted in adult patients. Alvarez et al. reported that comorbidities likely affect the outcome of SE in a relatively marginal way. The comorbidities of the adult patients in that study mainly included cerebrovascular disease. The presence of comorbidities does not necessarily predict a poor outcome [28, 29]. In our study, comorbid conditions (autism/hyperactivity/liver disease/renal disease/cerebrovascular disease) were detected in 46 (61%) of the total SE patients, and 26 (96%) of the SE patients had a poor prognosis. Comorbidities are important because of the contraindications and side effects of antiseizure drugs. In this regard, they may influence the outcome by influencing the treatment protocol.

STESS has been proven to be a good predictor of morbidity and mortality, and there is an urgent need for aggressive treatment. mSTESS was tested in a study in India by Sidharth et al. One hundred forty children (mean age: 5.8 years) were enrolled in this study. Seven children died, and 15 children overall had unfavorable outcomes. The predictive accuracy of the mSTESS was at a cutoff of ≥ 3 for unfavorable outcomes [11]. In our study, this cutoff was ≥ 2 for unfavorable outcomes and ≥ 4 for mortality. The difference of 1 point between the two studies could be due to the different groups of patients in the two studies. In our study, the mean age of the patients was 5.3 years, the percentage of male patients to female patients was almost the same, and the percentage of patients with previous epilepsy was 53%, which is similar to the variability observed in the study by Sidharth et al., except for the male/female ratio, which was greater in the study by Sidharth et al. Furthermore, the three most common SEs in our study were meningoencephalitis (19%), febrile status (8%), unknown (12%), and genetic causes (7%). In the study by Sidharth et al., the most common etiology was acute symptoms, which were present in 25.7% of patients, followed by remote symptoms, which were present in 25%, and febrile SE, which was present in 18.6%. The percentage of different etiologies between the two studies may play a role in producing different values for unfavorable outcomes.

5. Conclusion

The mSTESS is a useful and practical scoring system for predicting the prognosis of SE patients. Nonconvulsive SE, slowing and abnormal ground in EEG, being stuporous/comatose, abnormal MRI–CT, and the presence of comorbid conditions favored poor prognosis in SE patients in children.

Ethics Statement

The study was conducted in accordance with the ethical standards established in the Declaration of Helsinki (and its later amendments or comparable ethical standards) and approved by the Clinical Research Ethics Committee of Çukurova University (2021:2021/116).

Consent

Written informed consent was obtained from all of the participants and/or their legal guardians.

Disclosure

A preprint has previously been published in research square, and then, a reference to the preprint is included in the reference list [30].

Conflicts of Interest

The authors declare no conflicts of interest.

Author Contributions

S.B. planned the study; S.B. and G.G.M. coordinated the study; S.B., O.H., F.I., S.A., and G.G.M. analyzed the literature; and S.B., G.G.M., O.H., and F.I. analyzed the EEG collected data. D.Y. and O.O.H. followed patients in the intensive care unit, and D.K. conducted the psychomotor and developmental tests. S.B. was a major contributor to the writing of the manuscript study. All authors have read and approved the final manuscript.

Funding

There is nothing to declare. No funding has been received.