Abstract
Purpose :
Neurosurgical procedures may be considered potential confounding causes of retinal hemorrhage (RH) in children being evaluated for abusive head trauma. We sought to determine the prevalence and patterns of RH attributable to neurosurgical intervention in children.
Methods :
Retrospective cohort study of children undergoing neurosurgery who had postoperative indirect ophthalmoscopy within 7 days. Some children were also examined pre-operatively. Primary outcomes were prevalence and patterns of RH attributable to neurosurgery. Medical records were reviewed to identify confounding coexistent diseases.
Results :
573 eyes of 267 children (mean age 6.2 years, range 0.1–18), who underwent 289 neurosurgical procedures (101 craniectomy, 57 burr-hole related, 103 ventricular/cerebrospinal-fluid drain implants, 8 spinal surgery, and 20 other), were studied. Retinal examinations occurred at mean 3.1 days post-operatively (178 at 0-3 days, 111 at 4-7 days). RH’s were seen in 32 (11%) cases, but in every case they were either already present on pre-operative examination (13 cases) or matched the pattern of a co-existent known cause of RH, including head trauma with unambiguous history and non-ocular signs (13), hydrocephalus-related increased intracranial-pressure with papilledema-associated peripapillary RH (5), and ROP-ridge-associated RH (1). No RH could be attributed only to neurosurgery.
Conclusions :
Neurosurgery is unlikely to cause RH or be a significant confounding factor in the interpretation of retinal hemorrhage patterns in child abuse evaluations. While children undergoing child abuse evaluations may have intracranial hemorrhage requiring neurosurgery that occurs before a dilated retinal examination can be performed, our data suggest that neurosurgery itself is unlikely to produce RH.
This is a 2020 ARVO Annual Meeting abstract.