Systematic Evaluation of Radiation Dose Reduction in CT Studies of Body Packers: Accuracy Down to Submillisievert Levels.

Abstract

The objective of this study was to systematically evaluate the accuracy of abdominal CT performed at different radiation dose levels for the detection of body packs in human cadavers, in comparison with the accuracy of abdominal radiography. In this study, differing numbers of body packs (range, 0-20) were placed in the alimentary tract of human cadavers and then underwent imaging with abdominal radiography and with CT performed at different radiation dose levels (ranging from the standard abdominal CT dose to the technical minimum dose). Depiction of body packs on abdominal radiographs and on each CT scan was assessed by two independent blinded radiologists, and the accuracy of detection of body packs was calculated. The radiation dose associated with abdominal radiography was measured, and the effective radiation dose associated with CT was estimated. The mean (± SD) effective radiation dose for abdominal radiography was 1.4 ± 0.3 mSv, whereas the mean effective dose of CT ranged from 0.1 to 9.6 mSv. Interobserver agreement for body pack detection was moderate (κ = 0.51) for abdominal radiography and good (κ = 0.72-0.85) for CT. In a per-body pack analysis, abdominal radiography depicted 42% of the body packs with a sensitivity of 71% and a specificity of 100%. When performed at radiation dose levels of 0.6 mSv or greater, CT correctly detected all body packs. In per-person analysis, the sensitivity and specificity of CT for the correct detection of at least one body pack per cadaver was 100% for all radiation dose levels. CT performed at a dose of 0.6 mSv can be used for the detection of body packs. With a sensitivity and specificity of 100%, CT is superior to abdominal radiography in terms of reliability, associated radiation dose, and accuracy of detection.