Susceptibility-weighted imaging of retinal hemorrhages in abusive head trauma

Background Retinal hemorrhages are one of the most important supportive evidences for abusive head trauma (AHT). Susceptibility-weighted imaging (SWI) is highly suited to identify various forms of intracranial hemorrhage in AHT. However its utility in imaging retinal hemorrhage is not well establish...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Pediatric radiology 2019-02, Vol.49 (2), p.210-216
Hauptverfasser: Thamburaj, Krishnamoorthy, Soni, Ajay, Frasier, Lori D., Tun, Kyaw N., Weber, Sarah R., Dias, Mark S.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Background Retinal hemorrhages are one of the most important supportive evidences for abusive head trauma (AHT). Susceptibility-weighted imaging (SWI) is highly suited to identify various forms of intracranial hemorrhage in AHT. However its utility in imaging retinal hemorrhage is not well established. Objective SWI is a sensitive sequence for identifying retinal hemorrhage on MRI. Materials and methods In this retrospective analysis, 26 consecutive infants and young children with a suspected admission diagnosis of AHT underwent indirect ophthalmoscopy and brain MRI protocol for AHT along with SWI. Brain susceptibility-weighted images of 14 age-matched children were used as controls. For detecting retinal hemorrhage, susceptibility-weighted images of patients and controls were reviewed randomly and independently by two neuroradiologists who were blinded to the history and ophthalmology findings. A pediatric ophthalmologist graded the indirect ophthalmoscopy images. Results A diagnosis of AHT was confirmed in all 26 cases from a multidisciplinary meeting. Indirect ophthalmoscopy images were available in 21 cases. Ophthalmoscopy was positive for retinal hemorrhage in the right eye in 18 cases (85.7%) and in the left eye in 16 cases (76.2%). On SWI, retinal hemorrhage was identified in the right eye in 9/21 cases (42.8%) and in the left eye in 8/21 cases (38.1%) of AHT. Analysis of SWI in 21 cases of AHT demonstrated a sensitivity of 50%, specificity of 100%, positive predictive value of 100% and negative predictive value of 32% for detecting retinal hemorrhage. Conclusion SWI is moderately sensitive and highly specific for identifying retinal hemorrhage in AHT. Further studies are needed to identify steps to improve the efficiency of SWI in detecting retinal hemorrhage.
ISSN:0301-0449
1432-1998
DOI:10.1007/s00247-018-4292-8