Citation

Vinchon M, Defoort-Dhellemmes S, Desurmont M, Dhellemmes P. J. Neurosurg. 2005; 102(4 Suppl): 380-384.

Affiliation

Departments of Pediatric Neurosurgery, CHRU de Lille, Lille, France. m-vinchon@chru-lille.fr

Copyright

(Copyright © 2005, American Association of Neurological Surgeons)

DOI

unavailable

PMID

15926388

Abstract

OBJECT: Head injury is a major cause of morbidity and death in infants, and child abuse is among its chief causes. Retinal hemorrhages (RHs) are of paramount importance for the diagnosis of child abuse; however, their sensitivity and specificity are poorly estimated. Subdural hematoma (SDH) is a common feature; however, its incidence and causative factors are poorly documented. METHODS: To study the epidemiology of head injuries in infants, factors predisposing to SDH, value of RH for the diagnosis of child abuse, and prognostic factors, the authors prospectively collected data from cases of head injury in infants who were hospitalized at their institution over a 3-year period. One hundred fifty cases were collected, 57 of which were due to child abuse. This prospective study allows the comparison of clinical, radiological, and ophthalmological features in accidental and nonaccidental trauma in infants. CONCLUSIONS: Subdural hematomas were significantly correlated with RH and with child abuse but not with idiopathic macrocranium. The sensitivity and specificity of RH for the diagnosis of child abuse were 75 and 93.2%, respectively. Retinal hemorrhages associated with accidental trauma were always mild, and the specificity of more severe RH for the diagnosis of child abuse was 100%. The grading of RH requires the expertise of a trained neuroophthalmologist. Child abuse was also significantly associated with antecedents of perinatal illness, absence of signs of impact, and seizures on presentation. Although child abuse represented just 38% of traumas, it was the cause of 71% of deaths and 90% of severe disability in this series. Abuse and the clinical severity on presentation were two significant and independent factors conditioning outcome.