Periventricular hypodensity is associated with the incidence of pre-shunt seizure in hydrocephalic children.

Background: The seizure incidence in hydrocephalic children has been acknowledged in a lot of studies previously; nonetheless, seizure pathogenesis in these children remains unclear. Its high proportion of hydrocephalic children who underwent shunt surgery suggests that the seizure might be associated with the protocol of shunt placement and/or the shunt existence intracranially; however, this hypothesis could not explain the pre-shunt seizure incidence in hydrocephalic children.

Objective: This study aims to evaluate the patients' characteristics and CT findings in pre-shunt hydrocephalic children to identify the possible seizure etiology in these patients.

Methods: Three hundred and thirty-four children with hydrocephalus were included in this study, including 147 hydrocephalic children with the pre-shunt seizure history and 187 hydrocephalic children presented without the pre-shunt seizure history. The following information was retrieved from the patients' medical records: gender, age, pediatric Glasgow Coma Scale (pGCS) upon admission, and hydrocephalus diagnoses. CT findings were re-evaluated to assess the compression association of sulci and gyri, Sylvian fissures, cisterns, FH/ID ratio, Evan's ratio, and periventricular hypodensity with pre-shunt seizure.

Results: The results show that the pre-shunt seizure incidence is significantly higher in hydrocephalic children aged 1 to 5 years old (63/113 (55%), p = 0.0001), diagnosed with communicating hydrocephalus (97/163 (59%), p = 0.0001) or infectious hydrocephalus (80/109 (73%), p = 0.0001). The presence of periventricular hypodensity is significantly associated with the pre-shunt seizure incidence (132/205 (64.3%), p = 0.0001). Results from univariate analyses suggest significant association between periventricular hypodensity in every location and pre-shunt seizure (p < 0.0001). Multivariate analyses identify that temporal horn in the right lateral ventricle as the location of periventricular hypodensity has the strongest association with the pre-shunt seizure.

Conclusion: The presence of periventricular hypodensity in head CT scan is significantly associated with the pre-shunt seizure incidence. Further investigation to confirm this finding and evaluate the possible roles of inflammation in the pre-shunt seizure in hydrocephalic children is important to seek its possible implication on the treatment of pre-shunt seizure in these children.