BACKGROUND AND PURPOSE: MRI with sedation is commonly used to detect intracranial traumatic pathology in the pediatric population. Our purpose is to compare nonsedated ultrafast MRI (ufMRI), non-contrast head CT (nHCT), and standard MRI (stMRI) for detection of intracranial trauma in patients with potential abusive head trauma (AHT).
MATERIALS AND METHODS:A prospective study was performed in 24 pediatric patients who were evaluated for potential AHT. All patients received nHCT, ufMRI brain without sedation, and stMRI with general anesthesia or papoose, sequentially. Two pediatric neuroradiologists independently reviewed each modality blinded to other modalities for intracranial trauma. Inter-reader agreement was performed, and consensus interpretation for stMRI as the gold standard. Diagnostic accuracy was calculated for ufMRI, nHCT, and combined ufMRI with nHCT.
RESULTS:Inter-reader agreement was moderate for ufMRI (k=0.42), substantial for nHCT (k=0.63), and nearly perfect for stMRI (k=0.86). 42% of patients had discrepancies between ufMRI and stMRI which included detection of subarachnoid hemorrhage, and subdural hemorrhage. Sensitivity, specificity, positive and negative predictive values were obtained for any traumatic pathology for each exam: UfMRI (50%, 100%, 100%, 31%), nHCT (25%, 100%, 100%, 21%) and combination of ufMRI with nHCT (60%, 100%, 100%, 33%). UfMRI was more sensitive than nHCT for detection of intraparenchymal hemorrhage (p=0.03), and the combination of ufMRI with nHCT was more sensitive than nHCT alone for intracranial trauma (p=0.02).
CONCLUSION:In AHT, ufMRI, even combined with nHCT, demonstrated low sensitivity compared to stMRI for intracranial traumatic pathology which may limit its utility in this patient population.Abbreviations: AHT: abusive head trauma; GRE: gradient recalled echo; nHCT: non contrast head CT