Purpose or Case Report: Head trauma is a significant cause of morbidity and mortality in pediatrics. Non-contrast Head CT is the accepted gold standard imaging study to evaluate for suspected acute intracranial hemorrhage (ICH), however small acute extra-axial hemorrhage may be easily missed due to size and similar density to the bony calvarium. In Dual Energy CT (DECT), materials within the body such as bone/calcium and hemorrhage can be more easily discriminated based on differential attenuation at high and low peak voltage image acquisitions. This allows for advanced post-processing including automated bone removal which has been shown to improved visualization of acute ICH in the adult radiology literature, but has not yet been described in pediatrics. We report a retrospective review of DECT with automated bone removal for detection of acute ICH in the pediatric population.
Methods & Materials: Non-contrast head CTs ordered for “trauma” from the ED between 8/1/2018-8/1/2019 performed with DECT technique with final interpretation of “positive intracranial hemorrhage” were identified from the radiology information system (RIS). All CT exams were performed on a dual-source dual energy CT scanner (SOMATOM Force; Siemens Healthcare). Images available for review in PACS by interpreting radiologists included original DE datasets (80kV and Sn150 kV), mixed data set at a simulated 120 kV, and automated bone removal reconstructions (Dual Energy Bone Removal Application on syngo.via VA30A, Siemens Healthcare). CT images and both preliminary and final interpretations were reviewed retrospectively.
Results: A total of 114 intracranial hemorrhages are included in our study including 10 epidural hematomas (9%), 65 subdural hematomas (57%), 28 subarachnoid hemorrhages (25%), and 11 “extra-axial hemorrhage NOS” (10%). Of these, 57% (65/114) occurred after hours and had a formally documented preliminary interpretation by a radiology trainee, either 2nd year radiology resident or fellow. Detection of epidural and subdural hemorrhages was improved on DECT with automated bone removal. Overall rate of trainee discrepancy was similar to that reported in prior literature; however rate of discrepancy for subdural and epidural hemorrhages was improved.
Conclusions: This retrospective review highlights utility of DECT with additional automated bone removal for increasing conspicuity and detection of acute ICH in the pediatric population. To the best of our knowledge, this is the first report of this application in pediatrics.