CT of the chest in suspected child abuse using submillisievert radiation dose

The cornerstone of child abuse imaging is the skeletal survey, but initial imaging with radiographs may not demonstrate acute and non-displaced fractures, especially those involving the ribs. Given the high mortality of undiagnosed non-accidental trauma, timely diagnosis is crucial. CT is more sensitive in assessing rib fractures; however the effective radiation dose of a standard chest CT is high. We retrospectively identified four children (three boys, one girl; age range 1-4 months) admitted between January 2013 and February 2014 with high suspicion for non-accidental trauma from unexplained fractures of the long bones; these children all had CT of the chest when no rib fractures were evident on the skeletal survey. The absorbed radiation dose estimates for organs and tissue from the four-view chest radiographs and subsequent CT were determined using Monte Carlo photon transport software, and the effective dose was calculated using published tissue-weighting factors. In two children, CT showed multiple fractures of the ribs, scapula and vertebral body that were not evident on the initial skeletal survey. The average effective dose for a four-view chest radiograph across the four children was 0.29 mSv and the average effective dose for the chest CT was 0.56 mSv. Therefore the effective dose of a chest CT is on average less than twice that of a four-view chest radiograph. Our protocol thus shows that a reduced-dose chest CT may be useful in the evaluation of high specificity fractures of non-accidental trauma when the four-view chest radiographs are negative.