Objective: The objective of this study was to determine whether selected postmortem unenhanced CT findings can discriminate between antemortem and postmortem pericardial hemorrhage.
Materials and methods: Thirty-one consecutive cases of postmortem CT followed by autopsy identifying at least 50 mL of hemorrhaged blood were reviewed. Seven cases were classified as postmortem pericardial hemorrhage secondary to chest compression (postmortem group), and 24 cases were classified as antemortem pericardial hemorrhage secondary to disease or trauma (antemortem group), on the basis of autopsy findings. Postmortem CT findings of pericardial hemorrhage were classified as stratification comprising upper low-density and lower high-density areas (i.e., fluid-fluid level) and a high-density concentric ring (i.e., pericardial hyperdense ring). Diagnostic values for detecting antemortem or postmortem pericardial hemorrhage using this classification system, along with attenuation (in Hounsfield units) of pericardial hemorrhage lesions and the presence or absence of flattened heart sign on postmortem CT, were measured.
Results: There were statistically significant differences in fluid-fluid level and pericardial hyperdense ring between the postmortem and antemortem groups (p < 0.001). The sensitivity, specificity, positive predictive value, and negative predictive value of fluid-fluid level for detecting postmortem pericardial hemorrhage were 86%, 96%, 86%, and 100%, respectively. The mean attenuation of pericardial hemorrhage lesions differed statistically significantly between the groups (p = 0.004). The presentation of the flattened heart sign did not differ statistically significantly between the groups (p = 0.681).
Conclusion: Fluid-fluid level and pericardial hyperdense ring on postmortem CT, combined with attenuation, are useful for differentiating between antemortem and postmortem pericardial hemorrhage secondary to chest compression.
Keywords: MDCT; autopsy; cardiac tamponade; hemopericardium; postmortem diagnosis.