Retinal hemorrhage and brain injury patterns on diffusion-weighted magnetic resonance imaging in children with head trauma Gil Binenbaum, MD, MSCE, a,b Cindy W. Christian, MD, c,d Rebecca N. Ichord, MD, c,e Gui-shaung Ying, PhD, b Melissa A. Simon, MD, f Kathleen Romero, MD, g Avrum N. Pollock, MD, h,i and Brian J. Forbes, MD, PhD a,b PURPOSE To evaluate associations between retinal hemorrhage severity and hypoxic-ischemic brain injury (HII) patterns by diffusion-weighted magnetic resonance imaging (DW-MRI) in young children with head trauma. METHODS DW-MRI images of a consecutive cohort study of children under age 3 years with inflicted or accidental head trauma who had eye examinations were analyzed by two independent masked examiners for type, severity, and location of primary lesions attributable to trauma, HII secondary to trauma, and mixed injury patterns. Retinal hemorrhage was graded retro- spectively on a scale from 1 (none) to 5 (severe). RESULTS Retinal hemorrhage score was 3-5 in 6 of 7 patients with predominantly post-traumatic HII pattern and 4 of 32 who had traumatic injury without HII (P \ 0.001) on DW-MRI imag- ing. Severe retinal hemorrhage was observed in absence of HII but only in inflicted injury. Retinal hemorrhage severity was correlated with HII severity (r 5 0.53, P \ 0.001) but not traumatic injury severity (r 5 À0.10, P 5 0.50). HII severity was associated with retinal hemorrhage score 3-5 (P 5 0.01), but traumatic injury severity was not (P 5 0.37). CONCLUSIONS During inflicted head injury, a distinct type of trauma occurs causing more global brain injury with HII and more severe retinal hemorrhages. HII is not a necessary factor for severe retinal hemorrhage to develop from inflicted trauma. ( J AAPOS 2013;17: 603-608) U nderstanding of pediatric abusive head trauma has evolved over the last five decades. In 1962 Kempe and colleagues 1 published their land- mark paper “The Battered Child Syndrome,” the first real recognition of child abuse as a disease and of the responsibility physicians held for its diagnosis and pre- vention. The paper identified intracranial hemorrhage in young children as a hallmark sign in many cases. 1 Nine years later, in 1971, Guthkelch 2 was the first to suggest shaking as a form of abusive injury. He reported on 23 chil- dren (22 \18 months of age) presenting with various com- binations of subdural hemorrhage, fractures, parenchymal brain injury, and retinal hemorrhages. 2 Shortly thereafter, Caffey 3-5 coined the term whiplash-shaken infant syndrome. Both Guthkelch and Caffey noted a frequent absence of external signs of trauma and suggested the role of torn bridging vessels in the brain as the cause of the intracranial hemorrhage. Using autopsy evidence and a dummy model, Duhaime and colleagues 6 in 1987 sug- gested that blunt impact trauma may be a prerequisite to generate sufficient deceleration forces for the characteristic injuries to occur; however, consistency across perpetrator confessions suggests that shaking alone is sufficient to cause such injuries, 7 and actual injury threshold levels for infant brains have yet to be established. There are currently multiple hypothesized factors in the pathogenesis of brain pathology and retinal hemorrhage in abusive head trauma, including deceleration and sheering injury, hypoxic- ischemic injury (from decreased perfusion or apnea), blunt impact, neck flexion-extension, and raised intracranial or venous pressures. 8 However, the relative importance of these factors cannot be determined precisely based on the published data. The availability of diffusion-weighted magnetic reso- nance imaging (DW-MRI) and noninvasive vascular Author affiliations: Divisions of a Ophthalmology, c Pediatrics, and h Neuroradiology, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; Departments of b Ophthalmology, d Pediatrics, e Neurology, and i Radiology, Perelman School of Medicine and the University of Pennsylvania, Philadelphia, Pennsylvania; f Department of Ophthalmology, University of Medicine and Dentistry of New Jersey, Newark; g Department of Pediatrics, University of California San Diego, San Diego Supported by the National Institutes of Health grants K12 EY015398, P30 EY01583- 26, R01NS39679, and Loan Repayment Program. Presented in part at the 2011 Annual Meeting of the Association for Research in Vision and Ophthalmology in Fort Lauderdale, Florida, May 1-5. Submitted February 24, 2013. Revision accepted September 2, 2013. Published online November 9, 2013. Correspondence: Gil Binenbaum, MD, MSCE, Children’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., 9-MAIN, Philadelphia, PA 19104 (email: binenbaum@ email.chop.edu). Copyright Ó 2013 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 http://dx.doi.org/10.1016/j.jaapos.2013.09.002 Journal of AAPOS 603