Mediating effects of sleep quality between clinical characteristics and quality of life in children with epilepsy: A cross-sectional study from Southwest China

Abstract

BackgroundAlthough sleep quality (SQ) reportedly affects the health-related quality of life (QOL) of patients with epilepsy, little is known about the potential association between SQ and QOL, particularly in children with epilepsy (CWE). Our study aimed to investigate the mediating effect of SQ on the QOL of CWE to obtain more information for the prevention and treatment of epilepsy in children. MethodsWe collected general demographic and clinical data of 212 CWE and 79 controls (children who visited the Health Examination Department), and their guardians were instructed to answer the Children’s Sleep Habits Questionnaire (CSHQ) and the optimized Quality of Life in Childhood Epilepsy Questionnaire-16 (QOLCE-16). The t-test, analysis of variance, chi-square test, and Fisher’s exact test were used for between group comparisons. The Pearson correlation was used to analyze the correlation between variables. The direct, indirect, and total effects of predictors on the QOL of CWE were estimated based on an adjusted mediation model. ResultsCWE had significantly smaller long-term urban residence rates, less educated guardians, higher total CSHQ score, higher incidence of poor SQ, higher bedtime resistance, more sleep anxiety, worse sleep-disordered breathing, increased parasomnia, more daytime sleepiness, more frequent night waking, and greater sleep onset delay than controls (P < 0.05 for all). The univariable analysis showed significant differences in total CSHQ scores between CWE with different seizure frequency in the last month, whether or not drug-resistant epilepsy (DRE), and with different video electroencephalogram (VEEG) findings (P < 0.05 for all). Differences in QOLCE-16 scores between CWE with different guardian’s employment status, age at diagnosis, number of anti-seizure medication (ASM) types, seizure frequency in the last month, DRE status, seizure type, VEEG findings, neuropsychological evaluation findings, magnetic resonance imaging (MRI) findings, and etiology were statistically significant (P < 0.05 for all). The correlation study indicated that the total CSHQ score was negatively correlated with the QOLCE-16 score (P < 0.05). The mediation analysis showed that DRE and VEEG abnormalities had a standardized direct effect on the QOL. Seizure frequency in the last month, DRE, and VEEG abnormalities had an indirect effect on the QOL through SQ, and their mediating effect values of SQ were 31.61 %, 13.45 %, and 14.35 %, respectively. ConclusionOur findings uncovered the relationship of some clinical characteristics with SQ and QOL and characterized the nature of factors affecting the QOL of CWE. SQ could be a key factor in the prognosis of CWE experiencing epileptic seizures, and more attention should be paid on the management of SQ in interventions for epilepsy.