What neuroimaging should be performed in children in whom inflicted brain injury (iBI) is suspected? A systematic review
Received 3 August 2008; received in revised form 31 October 2008; accepted 11 November 2008.
Aims
To investigate the optimal neuroradiological investigation strategy to identify inflicted brain injury (iBI).
Materials and methods
A systematic review of studies published between 1970–2008 in any language was conducted, searching 20 databases and four websites, using over 100 keywords/phrases, supplemented by hand-searching of references. All studies underwent two independent reviews (with disagreements adjudicated by a third reviewer) by trained reviewers from paediatrics, paediatric neuroradiology and related disciplines, using standardized critical appraisal tools, and strict inclusion/exclusion criteria. We included primary studies that evaluated the diagnostic yield of magnetic resonance imaging (MRI), in addition to initial computed tomography (CT), or follow-up CT or ultrasound in children with suspected iBI.
Results
Of the 320 studies reviewed, 18 met the inclusion criteria, reflecting data on 367 children with iBI and 12 were published since 1998. When an MRI was conducted in addition to an abnormal early CT examination, additional information was found in 25% (95% CI: 18.3–33.16%) of children. The additional findings included further subdural haematoma, subarachnoid haemorrhage, shearing injury, ischaemia, and infarction; it also contributed to dating of injuries. Diffusion-weighted imaging (DWI) further enhanced the delineation of ischaemic changes, and assisted in prognosis. Repeat CT studies varied in timing and quality, and none were compared to the addition of an early MRI/DWI.
Conclusions
In an acutely ill child, the optimal imaging strategy involves initial CT, followed by early MRI and DWI if early CT examination is abnormal, or there are ongoing clinical concerns. The role of repeat CT imaging, if early MRI is performed, is unclear, as is the place for MRI/DWI if initial CT examination is normal in an otherwise well child.
aDepartment of Child Health, Wales School of Medicine, Cardiff University, Cardiff, UK
bDepartment of Child Health, Sue Nicholls Centre, Aylesbury, UK
cSupport Unit for Research Evidence, Cardiff University, Cardiff, UK
dDepartment of Primary Care & Public Health, Cardiff University, Cardiff, UK
eDepartment of Neuroradiology, Wessex Neurological Centre, Southampton University Hospitals Trust, UK
fImaging Centre, University Hospital, Nottingham, UK
Guarantor and correspondent: S. Maguire, Department of Child Health, Wales School of Medicine, Cardiff University, Health Park, Cardiff, South Glamorgan, CF 14 4 XN, UK. Tel.: +44 2920 742160; fax: +44 2920 744283.
ABSTRACT