1 Introduction

Herpes simplex encephalitis (HSE) is viral encephalitis caused by herpes simplex virus (HSV) infection and is a common form of sporadic encephalitis [1]. There is no clear seasonal pattern and it occurs equally frequently in males or females, with a peak incidence at 6 months to 3 years of age, and about 1/3 of patients are from 6 months to 20 years of age [2,3,4]. The introduction of acyclovir (ACV) therapy has significantly reduced mortality in HSE, but epilepsy and potential neurological sequelae such as intellectual disability, psychomotor retardation, speech, and memory impairment are common in young patients [5, 6].

Epilepsy is one of the major sequelae after the HSE cure, which may be caused by sequelae of irreversible neurological damage or by persistent viral or immune inflammation in the brain [7]. In addition, seizures can lead to neuronal loss in the hippocampus, amygdala, cerebral cortex, thalamus, cerebellum, and other parts of the brain [8], the effects on young children during their neurological development are enormous [9, 10].

Therefore, children with HSE need to be closely monitored to prevent post-encephalitic epilepsy (PE) after they are cured. This study analyzed the clinical features of pediatric patients with HSE and explored the risk factors for PE to provide a basis for clinical management or prognostic assessment of children with HSE.

2 Patients and methods

2.1 Patients

Using the electronic medical record system of the First Affiliated Hospital of Zhengzhou University, the search dates were set between January 2013 and June 2021, and the age of the patients was set under 18 years old. The following terms “HSV encephalitis”, “herpes simplex encephalitis”, “herpes simplex virus encephalitis”, and “herpes encephalitis” were searched in the system, and forty-seven children who met the diagnostic criteria for HSE were manually screened and enrolled in our study, and their clinical data were collected.

The following diagnostic criteria were established based on the literature [11,12,13] and the normal range of laboratory tests at the First Affiliated Hospital of Zhengzhou University: encephalitis was defined as altered mental status (reduced or altered level of consciousness, drowsiness, or personality changes) lasting more than 24 h with at least three of the following associated manifestations: (1) temperature ≥ 38 °C within 72 h before and after the visit, (2) generalized or focal seizures, (3) new onset of focal neurological findings, (4) cerebrospinal fluid (CSF) white blood cell count ≥ 15×106/L, and (5) abnormal EEG. The diagnostic criteria for HSE met the above criteria for encephalitis and at least one of the following biological criteria: (1) HSV nucleic acid detected by polymerase chain reaction (PCR) in CSF or brain tissue; (2) HSV-specific antibodies detected in CSF; (3) a four-fold increase in HSV-specific IgG antibody levels in 2 CSF samples and/or 2 blood samples at week 1 of symptom onset and 3 weeks after onset. Patients with other neurological diseases such as cerebral hemorrhage, cerebral ischemia, toxic encephalopathy, metabolic encephalopathy, intracranial tumor, autoimmune encephalitis, etc. were excluded.

Inclusion Criteria: (1) meeting the diagnostic criteria for HSE; (2) normal neurological development and no history of seizures before the diagnosis of HSE; (3) the patient's age ≤ 18 years.

Exclusion Criteria: (1) metabolic and/or toxic encephalopathy, autoimmune encephalitis, or other non-CNS infectious diseases; (2) severe cardiac, pulmonary, hepatic, or renal dysfunction as well as acid–base metabolic disorders; (3) previous history of cerebrovascular disease, brain tumor, trauma, surgery, or epilepsy; (4) immunodeficiency diseases; (5) less than six months of follow-up, missed follow-up, or family refusal to follow up.

A total of 47 children, one case was excluded due to combined immunodeficiency and one case was excluded due to loss of follow-up, and finally 45 children were included in the trial. All children with HSE were followed up by outpatient or telephone for > 6 months, once a month for the first 3 months, then every 3 months, and every 6 months after 1 year. According to whether the last follow-up was PE or not, they were divided into PE and non-PE groups. Diagnostic criteria for PE were established based on the relevant literature [14, 15] and the International League Against Epilepsy(ILAE) [16] epilepsy definition: with at least 2 non-evoked seizures (> 24 h interval) or oral anti-epileptic drugs (AEDs) treatment for > 6 months after the acute phase of HSE, with a re-examination of cranial imaging and/or EEG still abnormal for epileptic seizures.

2.2 Clinical Data

Collected the following information from children in 2 groups: (1) age, gender; (2) fever duration, fever peaks; (3) psychiatric behavior disorders; (4) neurological deficit symptoms and/or signs (aphasia, hemiparesis, hemianesthesia, hemianopsia, etc.); (5) altered consciousness: (somnolence, confusion, lethargy, delirium, and coma); (6) type of epileptic seizures in the acute phase (focal, generalized), seizures and seizure frequency in the acute phase (HSE within 7 days of the onset of symptoms or signs[15, 17]), status epilepticus. (7) Glasgow coma score (GCS) on admission, admission to intensive care unit (ICU) or not, ICU length of stay; (8) cranial imaging (CT/MRI/PET-CT): completion, normal, abnormal (cortical involvement, subcortical involvement, mixed involvement); (9) electroencephalogram (EEG): completion, normal, abnormal; (10) cerebrospinal fluid (CSF) cell count: normal, increased (> 15 × 106/L) and values. Data is collected by two people at the same time, checked and collated after collection. The process of collection is anonymized, and children's names were recorded by numerical numbers.

2.3 Statistical Analysis

Quantitative data are presented as \(\overline{x} \pm s\) when normally distributed, using two independent-samples t-test; skewed distribution is presented as M (P25, P75), using U-test. Statistical data were expressed as percentages (%) using the χ2 test or Fisher’s exact test. Multivariate Logistic regression analysis was used to analyze statistically significant items for single-factor analysis. Every test was two-sided, and a statistically significant difference was defined as P < 0.05. Software called SPSS 22.0 was used to do the statistical analysis.

3 Results

3.1 Clinical Features

The demographics and clinical symptoms of children with HSE at admission are shown in (Table 1). Of the 45 children with HSE, 22 (48.9%) were male and 23 (51.1%) were female, with a roughly even gender distribution; the youngest child was 8 months old and the oldest 192 months (16 years), with a median age of 69 months. Among the initial symptoms, there were 40 (88.9%) cases of fever, of which 13 (32.5%) showed fever only, 13 cases (32.5%) of fever with a headache, 12 cases (30%) of fever with convulsion, 1 case (2.5%) of fever with dizziness, and 1 case (2.5%) of fever with limb weakness. There were 15 cases (33.3%) of convulsions, of which 3 (20%) showed only convulsions; 15 cases (33.3%) of headaches, of which 2 (13.3%) showed only headaches, and most children had two or more initial symptoms.

Table 1 Clinical features of HSE in children
Full size table

Other symptoms in the acute phase included seizures in 26 cases (57.8%), psychiatric behavioral abnormalities in 13 cases (28.9%), neurological deficit symptoms in 16 cases (35.6%), altered consciousness in 24 cases (53.3%), and admission to the pediatric ICU in 31 cases (68.9%). After admission, cerebrospinal fluid cell count was abnormal in 34 cases (75.6%); EEG was completed in 31 cases (68.9%), of which 24 cases (77.4%) were abnormal; and cranial imaging was completed in 38 cases (84.4%), of which 34 cases (89.5%) were abnormal, with higher implementation rate and sensitivity of cranial imaging.

During medication and follow-up,26 cases (53.3%) of children with seizures in the acute phase and 24 cases (53.3%) on oral AEDs, while 25 cases (55.6%) of oral AEDs after the acute phase, with a minimum of 1 month and a maximum of 18 months of cumulative oral AEDs and a median cumulative duration of 7 months on oral AEDs. The median follow-up time was 10 months (7–23 months), and 19 cases (42.2%) developed PE by the end of the follow-up (Table 1).

3.2 Risk Factors of PE in Children With HSE

The PE and non-PE groups of patients with HSE showed statistically significant differences in fever duration (P = 0.008), fever peaks (P = 0.014), psychiatric behavior disorders (P = 0.019), neurological deficit symptoms(P = 0.041), acute phase seizures and several seizures (P = 0.002 and P < 0.001), ICU length of stay (P = 0.015), admission GCS score (P = 0.002), and cranial imaging indicating mixed involvement (P = 0.020, fisher’s exact test) compared (Table 2).

Table 2 Comparison of clinical features of children with HSE between the PE group and non-PE group
Full size table

Statistically significant variables from the univariate analysis were included in the multivariate logistic regression analysis, after automatic and manual “enter” screening in SPSS, the seizure frequency in the acute phase was retained as significant in the model (OR = 1.551; 95% CI 1.229–1.958, P < 0.001), and the ROC (receiver operating characteristic) curve was applied, resulting in a Youden index = 0.650, corresponding to a cutoff = 5.5. Thus, repetitive seizures (more than 5 seizures) in the acute phase was an independent risk factor for the development of PE in children with HSE (Table 3).

Table 3 Multivariate Logistic regression analysis of PE for children with HSE
Full size table

4 Discussion

HSE is one of the most severe clinical presentations of HSV infection, with children being the victims in about one-third of incidents. The signs and symptoms of HSE reflect the inflammatory reaction and the virus's proliferation in brain tissue [18]. Some cases may also cause autoimmune encephalitis [19]. Common clinical symptoms include fever, headaches, psychiatric behavioral problems, and seizures [1, 20, 21].

The initial symptoms of the children with HSE in this study were fever, headache, and convulsions, and 27 (60.0%) cases had fever accompanied by headache and convulsions. The acute phase is primarily characterized by symptoms such as fever, seizures, altered consciousness, neurological deficit symptoms, and psychiatric behavioral abnormalities. Forty cases (88.9%) had a fever in the acute phase, 26 cases (57.8%) had seizures, and 24 cases (53.3%) had altered consciousness, all of which were similar to the proportions in the study by Hsieh, W B [22]. The proportion of acute phase seizures in some of the relevant studies [2, 5] was concentrated around 70%, but the number of subjects included in their studies was all smaller, and relevant studies with larger numbers of participants are still needed to prove this. Headache is a common complaint in adults with HSE, but only 15 cases (33.3%) presented with headaches in the current study of pediatric HSE, a similar condition reported in previous studies [22]. In this study, there were 16 cases (35.6%) with neurological deficit symptoms, similar to those reported by Sili et al. [23].

It has been shown that nearly two-thirds of patients with HSE will have significant residual neurological deficits, leading to a poor long-term neurological prognosis, even if they are treated appropriately early in the disease episode [2, 5]. In this study, the incidence of PE in children with HSE was as high as 42.2% (19/45 cases) at 7 to 23 months after encephalitis, which is similar to the incidence in previous related studies [5, 13, 24], indicating that children with HSE have a higher incidence of PE in the future.

Whether pediatric HSE transforms into PE and what are the risk factors for transformation into PE are important for the pharmacological treatment as well as prevention of children with HSE. Previous studies have found that PE is associated with status epilepticus, acute phase seizures, structural cranial imaging lesions, ICU admission, and abnormal EEG discharge activity, with status epilepticus as an independent risk factor for PE [25, 26]. However, there are few reports on the factors influencing the occurrence of PE in pediatric patients with HSE. In this study, 45 children with HSE were followed up for more than 6 months, and univariate analysis revealed that fever duration, fever peaks, psychiatric behavior disorders, neurological deficit symptoms, acute phase seizures and seizure frequency, ICU length of stay, admission GCS score, and cranial imaging indicating of mixed involvement were associated with the occurrence of PE in HSE children. Of these, acute phase seizures, cranial imaging abnormal, and ICU length of stay were similar to those reported previously [26, 27], but the EEG abnormalities in this study were different from those they had reported, probably because only children with HSE were enrolled in this study, for reasons to be further investigated (Table 2). Multivariate logistic regression analysis revealed that repetitive seizures (more than 5 seizures) in the acute phase was an independent risk factor for the development of PE in children with HSE (Table 3), therefore, long-term application of AEDs is recommended for children with HSE with the above conditions, especially multiple conditions.

5 Conclusion

The main symptoms in children with HSE in the acute phase include seizures, altered consciousness, neurological deficit symptoms, and psychiatric behavioral abnormalities. Children with HSE has a high incidence of PE. In the acute phase, one or more risk factors occur, especially if they have repetitive seizures (more than 5 seizures), long-term application of AEDs should be started as early as possible to prevent the development of PE.

There are some limitations of this study: 1. This study is a single-center retrospective study with a small sample size, which may be the reason why there was only 1 independent risk factor in this study at the time of multivariate logistic regression analysis. The sample size should be expanded as much as possible in the follow-up study to conduct a multicenter, prospective study.2. Because the number of seizures in the acute phase was recorded within 7 days, we used only the number of seizure symptoms that appeared clinically after the child was diagnosed with acute epilepsy as the number of seizures, which may lead to bias in the number of seizures in the acute phase, and subsequent studies should try to capture seizures recorded by EEG.