Citation

Minns RA, Jones PA, Fleck BW, Mulvihill AO. Child Abuse Rev. 2022; 31(1): 54-65.

Copyright

(Copyright © 2022, John Wiley and Sons)

DOI

10.1002/car.2707

PMID

unavailable

Abstract

One important cause of retinal haemorrhages (RHs) in children is abusive head trauma (AHT). Our programme of research was conducted on children admitted to hospital over a seven-year period who had RHs as part of their clinical presentation. This paper summarises important findings from previous publications. As a prerequisite, two basic studies were necessary: (i) a method of reporting the precise location of the RH in the retina; and (ii) defining a simple clinical classification of RHs based on the appearance of the retinal layer involved. Because raised intracranial pressure (RICP) is a frequent secondary brain insult following serious adult head injury, causing hypoxic-ischaemic injury with a resultant poor outcome, it was considered whether such RICP might also cause RHs in children. Detailed, state-of-the-art monitoring methods were used to investigate this relationship along with retinal imaging.

METHODical documentation of the type, area and lifespan of RHs meant that it was possible to address the question of predicting AHT from RH numbers and characteristics, and to describe not only the duration of different layer haemorrhages, but also a previously unreported observation that some RHs transiently enlarge over two to three days before resolving normally. 'One important cause of retinal haemorrhages (RHs) in children is abusive head trauma' Key Practitioner Messages A new retinal zone classification for research and legal purposes has been developed, as well as a simplified five-point working classification of RHs, based on existing fundoscopic descriptions. In this cohort study, a young age and a high 'dot-blot' count (>25 intraretinal haemorrhages) are strong predictors of AHT. There is a complex association between the burden of intracranial pressure insult and RHs. Time to resolution of different RHs has been calculated using two-dimensional RH area pixel counts from sequential retinal imaging. Two patterns of resolution are reported.

Language: en

Keywords

abusive head trauma; assessment; cohort study; non-accidental head injury; physical abuse