Factors Affecting the Recurrence of Seizure in Children with Developmental Disorders Using Frailty Model for Recurrent Event and Considering Restraint Period

 Seizure is the most common neurological disorder in pediatric medicine. This disorder is often chronic and requires continuous treatment. Developmental disorders are conditions that can occur due to abnormal development or damage to the brain and central nervous periods (out-of-hospital). However, in most studies, the severity of seizure relapse is mistakenly considered the same for each period. The present study aimed to identify  factors related to the recurrence of seizure in children with developmental disorder.

In this retrospective cohort study, 228 children with developmental disorder and a history of seizures referred to the emergency department of the Children’s Hospital Medical Center in Tehran, Iran in 2016. Relapse times, admission duration and discharge time, age, sex, birth weight, type of delivery, parental relatives, maternal age during pregnancy, mother's history of medication use and disease during pregnancy, history of seizure with fever, family history of seizure, maternal delivery problems, length of pregnancy, history of admission to neonatal intensive care unit were recorded as factors affecting recurrence of seizure. Inclusion criteria were having developmental disorder and a history of seizure recurrence. The patients who were hospitalized for any reason other than seizure were excluded from the study. No intervention was performed in this study. Patients' information was recorded with their permission and based on ethical principles and the results were presented in the form of tables. Survival model was used to investigate the effect of risk factors on the seizure recurrence. If the seizure recurrence in children was controlled in hospital by medical staff, response variable (relapse time) was considered as censored. A frailty parameter was used to explain the correlation between recurrence of seizures in one subject, and a correlation parameter was used to justify the difference in in the rate of correlation between seizures in a hospital and in somewhere other than hospital. The collected data were analyzed using frailty model for recurrent event by considering restraint period. To achieve an effective and flexible approximation for estimating base intensities, the piecewise constant intensity method was used. Estimation of the parameters was performed using the Gaussian quadrature method and by NLMIXED procedure in SAS v. 9.2 software.

 Of 228 children, 125 (54.8%) were boys and half of them were under 3 years of old. The mothers of 200 children (87.72%) had age of 18-35 years at the time of pregnancy; 96 (41.67%) of these patients were admitted to neonatal intensive care unit during neonatal period. The birth weight of 199 (28.27%) children was about 2.5-4 kg and 119 did not experience any seizure during the restraint period (in the hospital). The frequency of seizure recurrence in hospital ranged from 1 (25.88%) to 16 times (0.44%), while in out-of-hospital, it ranged from 1 (7.46%) to 11 times (0.44%). The duration of seizure recurrence was between 1 to 6196 days. Furthermore, children aged 1-2 years (P<0.001) with a history of admission to neonatal intensive care unit (P<0.001) whose mothers were 18-35 years of old during pregnancy (P=0.022) experienced more out-of-hospital seizure recurrence, while children aged 1-2 years (P=0.007) with a birth weight of 2.5-4 kg (P= 0.037) experienced less seizure recurrence in hospital.

 Considering the meaningful relationships and appropriate planning, the recurrence of seizure can be reduced in children with developmental disorder during hospitalization or after withdrawal.