Echocardiogram and clinical features help researchers identify epilepsy after encephalitis in children.
Early prediction of epilepsy after encephalitis in childhood can be identified by the peripheral blood hemoglobin, the number of epileptic seizures in the acute stage of encephalitis, and electroencephalogram (EEG) slow wave/fast wave frequencies, a study finds.
“In this study, 46 of the 255 children with encephalitis were diagnosed with postencephalitis epilepsy; that is, the incidence was 18.04%, which is consistent with the previously reported incidence,” wrote the researchers of the study.
The single center retrospective study is published in Emergency Medicine International.
Encephalitis is an infectious disease that affects the central nervous system and can result in seizures. Although most patients with encephalitis can control their seizures with treatment, some patients continue to have seizures after the cure of encephalitis, requiring long-term use of antiepileptic drugs to control.
EEG is an important tool in the diagnoses of epilepsy, determining seizure type, and determining epilepsy syndrome. The study aimed to evaluate an early prediction model of epilepsy after encephalitis in childhood based on echocardiogram (ECG) slow wave/fast wave frequency and clinical features.
From May 2019 to December 2021, the researchers enrolled a total of 255 children with encephalitis at the Second Affiliated Hospital of Nantong University who were aged 3 to 12 years. Eligibility criteria included having epilepsy in the acute stage of encephalitis and follow-up for at least 1 year after discharge.
Exclusion criteria were complications with malignant tumor, chronic infectious diseases, heart disease, autoimmune disease; sever liver and kidney dysfunction; history of epilepsy; loss of connection or death at follow-up; incomplete clinical data, unsigned informed consent.
Patients were randomly divided into training and verification sets, as well as the presence of postencephalitic epilepsy (PE) and no postencephalitic (no-PE) depending on whether epilepsy occurred 1 year after discharge. In the data analysis, the researchers used univariate and multivariate logistic regression to create a prediction model in the training set and evaluated the performance of the model in the diagnosis of epilepsy in patients with cognitive functions.
Of these patients, there were 184 in the training set and 71 in the verification set. Additionally, 209 patients were in the non-PE group and 46 patients in the PE group.
The significant findings found showed that hemoglobin (OR, 0.97; 95% CI, 0.95-0.96), epilepsy frequency (OR, 0.97; 95% CI, 0.95-0.96), and ECG slow wave/fast wave frequency in the occipital region were independent influencing actors for PE (P < .05).
The prediction model is based on the above factors: -0.031 × hemoglobin -2.113 × epilepsy frequency + 7.836 × occipital region S/F + 1.595.
Furthermore, the areas under the receiver operating characteristic curve for the diagnosis of PE in the training set and the validation set were 0.835 and 0.712, respectively.
The researchers acknowledge some limitations to the study, including being in a single center, had a low sample size, and was restricted by regional location.
Despite these limitations, the researchers believe these findings suggest the sensitivity and specificity of the PE prediction model was high. Furthermore, the study suggests that blood hemoglobin, the number of epileptic seizures in the acute stage of encephalitis, and EEG slow wave/fast wave frequencies can be used as predictors for seizures in children with epilepsy after encephalitis.
Reference
Sun X, Zhao J, Guo C, Zhu X. Early prediction of epilepsy after encephalitis in childhood based on EEG and clinical features. Emerg Med Int. 2023;2023:8862598. doi:10.1155/2023/8862598
ATTR-CA Variant Disproportionately Impacts Black Individuals With Heart Failure
September 16th 2025The transthyretin cardiac amyloidosis (ATTR-CA) variant is more prevalent in Black individuals, increasing the need for more research to assess hereditary links and early prevention awareness.
Read More
The Breakdown: Breast Cancer Research Awareness Day
August 19th 2025Breast cancer is the second most common cancer among women and the second leading cause of cancer-related deaths among women in the US. In light of Breast Cancer Research Awareness Day, The American Journal of Managed Care® breaks down the most recent advancements in breast cancer prevention, screening, and therapies.
Listen
The Future of Survivorship Care: Precision Tools for Women After Breast Cancer
September 10th 2025The Breast Cancer Index and artificial intelligence–assisted mammograms are tools that can help mitigate symptoms and reassure patient insecurities of women with breast cancer post-treatment.
Read More
Racial, Ethnic Disparities May Impact Age Differences at First Heart Failure Hospitalization
September 4th 2025Social determinants of health were associated with a younger age of first hospitalization for heart failure in Black and Hispanic patients when compared with Asian and White patients.
Read More