Sensitivity of autopsy and radiological examination in detecting bone fractures in an animal model: implications for the assessment of fatal child physical abuse.

Skeletal injuries are often strong indicators of child abuse and their detection is therefore crucial. The aim of this study was to compare the sensitivity of three diagnostic approaches, namely autopsy, traditional (conventional) radiology, and computed tomography on "battered" piglets, in order to verify the sensitivity of each method, with respect to the true number of bone fractures assessed once the piglet was skeletonised (osteological control). Four newborn cadaver piglets who had died from natural causes were severely beaten post-mortem in every district of the body. Traditional radiography, computed tomography (CT) and autopsy were performed. The piglet was then macerated until skeletonised and the number of all fractures present recorded (osteological control). On the cranium, traditional radiology revealed only 35% circa of actual fractures, autopsy detected only 31% (P<0.01 for both comparisons versus osteological control), whereas CT imaging detected all fractures actually present. For ribs, radiology detected only 47% of all fractures present, and autopsy 65% circa (P>0.05 for both comparisons versus osteological control), while CT scans detected 34% (P<0.01). In suspected cases of fatal child abuse, we suggest that the bones of specific districts be directly analysed either at autopsy or by collecting specific diagnostic sites, such as parts of the rib cage, and subjecting them to maceration. The removed areas could be replaced with artificial material for cosmetic purposes. The authors stress the importance of combined radiological, CT scan, autopsy and osteological survey in the detection of perimortem bone fractures.