David T F Sun1, X L Zhu, W S Poon. 1. Division of Neurosurgery, Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, New Territories East, Hong Kong.
Abstract
OBJECTIVE: We aim to identify the incidence and the characteristics of non-accidental subdural haemorrhage in Hong Kong in children under the age of 5 years. METHOD: Eighteen children aged below 5 years presented with subdural haemorrhage without a history of significant trauma were studied. Clinical features and the work-up process for probable child abuse were analysed. RESULTS: Sixteen combined case conferences were held, and eleven cases were concluded to be genuine child abuse cases of non-accidental head injury (incidence=1.5 per 100,000 children <5 years old per year). Among these cases, the most common presentation in the younger age group (age<1 year) was seizure (5/7), whilst that in the older age group (age 1-4 years) was coma (4/4, two of these patients also had seizure). One patient in the younger age group and three (3/4) of the older age group were found to have multiple unexplained bruises, suggesting that, in older children, direct impact injury played an important role. Five (5/7) children in the younger age group and all four children in the older age group had retinal haemorrhage. Coma on admission was associated with an unfavourable outcome (6/7 vs 0/4, p=0.015). CONCLUSIONS: Sixty-one percent of childhood subdural haemorrhage (11/18) without a history of significant trauma were genuine child abuse cases and should all be investigated for probable "non-accidental injury". An organised protocol involving a multi-disciplinary team is mandatory.
OBJECTIVE: We aim to identify the incidence and the characteristics of non-accidental subdural haemorrhage in Hong Kong in children under the age of 5 years. METHOD: Eighteen children aged below 5 years presented with subdural haemorrhage without a history of significant trauma were studied. Clinical features and the work-up process for probable child abuse were analysed. RESULTS: Sixteen combined case conferences were held, and eleven cases were concluded to be genuine child abuse cases of non-accidental head injury (incidence=1.5 per 100,000 children <5 years old per year). Among these cases, the most common presentation in the younger age group (age<1 year) was seizure (5/7), whilst that in the older age group (age 1-4 years) was coma (4/4, two of these patients also had seizure). One patient in the younger age group and three (3/4) of the older age group were found to have multiple unexplained bruises, suggesting that, in older children, direct impact injury played an important role. Five (5/7) children in the younger age group and all four children in the older age group had retinal haemorrhage. Coma on admission was associated with an unfavourable outcome (6/7 vs 0/4, p=0.015). CONCLUSIONS: Sixty-one percent of childhood subdural haemorrhage (11/18) without a history of significant trauma were genuine child abuse cases and should all be investigated for probable "non-accidental injury". An organised protocol involving a multi-disciplinary team is mandatory.