Experience with post-mortem computed tomography in Southern Denmark 2006-11

Objectives: (1) To explore the ability of post-mortem computed tomography (PMCT) to establish the cause of death. (2) To investigate the inter-method variation between autopsy and PMCT. (3) To investigate whether PMCT can select cases for autopsy. (4) To investigate the importance of histology.
Materials: PMCT and autopsy were performed in 900 forensic cases from Southern Denmark. 4547 diagnoses were registered.
Methods: This was a prospective, double-blind investigation.
Results: In twothirds of all cases, PMCT and autopsy agreed on the cause of death. The agreement was highest for injury deaths and lower for deaths caused by disease. Twothirds of the diagnoses were obtained by both autopsy and PMCT. PMCT reliably detected air and fluid collections, hyper-and hypotrophy, fractures, neoplasms, gallstones and kidneystones, calcification, foreign bodies, aneurisms, hernias and cerebral hemorrhages and was less reliable at detecting injuries in inner organs, small hematomas and vesseltransections. PMCT was unable to detect thrombosis, thromboembolism, cirrhosis, esophagealvarices, myocardialinfarctions and myocardial scartissue, peritonitis and gastro-intestinal ulcerations. At least 15% of the autopsies in this study could have been substituted by PMCT. Histology yielded important new information in onefourth of the cases, but this proportion was much smaller among cases for which we estimated that the autopsy could have been substituted by PMCT. Conclusion: PMCT is a valuable supplement to autopsy and may be used as a screeningtool for selecting cases for autopsy. International standardization of interpretation, reporting and coding of autopsy and PMCT findings are encouraged.