Postmortem CT is more accurate than clinical diagnosis for identifying the immediate cause of death in hospitalized patients: a prospective autopsy-based study

Despite 75 to 90 % physician accuracy in determining the underlying cause of death, precision of determination of the immediate cause of death is approximately 40 %. In contrast, two thirds of immediate causes of death in hospitalized patients are correctly diagnosed by postmortem computed tomography (CT). Postmortem CT might provide an alternative approach to verifying the immediate cause of death. To evaluate the effectiveness of postmortem CT as an alternative method to determine the immediate cause of death in hospitalized patients, an autopsy-based prospective study was performed. Of 563 deaths from September 2011 to August 2013, 50 consecutive cadavers undergoing hospital autopsies with consent for additional postmortem CT at the University of Fukui were enrolled. The accuracy of determination of the immediate cause of death by postmortem CT was evaluated in these patients. Diagnostic discrepancy was also compared between radiologists and attending physicians. The immediate cause of death was correctly diagnosed in 37 of 50 subjects using postmortem CT (74 %), concerning 29 cases of respiratory failure, 4 of hemorrhage, 3 of liver failure and 1 of septic shock. Six cases of organ failure involving 13 patients were not identified as the cause of death by postmortem CT. Regarding the immediate cause of death, accuracy of clinical diagnosis was significantly lower than that of postmortem CT (46 vs 74 %, P < 0.01). Postmortem CT may be more useful than clinical diagnosis for identifying the immediate cause of death in hospitalized patients not undergoing autopsy. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s00428-016-1937-6) contains supplementary material, which is available to authorized users.