Citation

Soto-Ares G, Denes M, Noulé N, Vinchon M, Pruvo JP, Gosset D. J. Radiol. 2003; 84(11 Pt 1): 1757-1765.

Vernacular Title

Collections sous durales de l'enfant: place de l'IRM cerebrale et medullaire dans

Affiliation

Service de Neuroradiologie, CHRU de Lille, 59037 Lille Cedex. gsotoares@chru-lille.fr

Copyright

(Copyright © 2003, Elsevier Publishing)

DOI

unavailable

PMID

15022989

Abstract

PURPOSE: To determine if the presence of multiple subdural hematomas (SH) of varying signal intensities at MRI was suggestive of repeated injuries allowing a diagnosis of non accidental head injury (NAHI). MATERIAL AND METHODS: We included 13 patients (9 boys/4 girls; mean age 15.3 weeks) with suspected NAHI and bilateral SH. Six patients had MRI during acute admission (first week) and seven patients between 10 and 30 days after head trauma (mean 12.6 days). A total of 8/13 patients had spinal MRI. We assessed the distribution and signal intensity of SH, brain, epidural and spinal cord abnormalities. Our findings were compared to medico-legal investigations in 7/13 patients to evaluate the correlation between the age of SH on MRI and the date of injuries. RESULT: Lesions detected included cerebral edema, ischemia, contusions and cranio-cervical axonal injury and 2/8 patient had spinal epidural hematomas. Multiple SH with varying signal intensities were seen. The commonest location was the frontoparietal areas (100%) and suboccipital region (77%). In 5/6 cases, MRI during the acute admission revealed multiple and small SH having the same signal intensity. When MRI was performed later multiple SH of varying signal intensity were found in 54% of cases. We found a good correlation between the age of the SH and the date of suspected injuries in all cases. Nevertheless, in 5 patients, other SH of various ages were present suggesting rebleeding. CONCLUSION: Our findings confirm the value of MRI in the detection of brain and spinal cord lesions and SH of varying signal intensities in non accidental head injury. MRI is not able to predict whether different signal intensities correspond to spontaneous repeated bleedings or repeated injuries. The presence of these lesions without clinical evidence of accidental trauma justifies a medico-legal evaluation to confirm the diagnosis of child abuse and to protect these patients.

Language: fr