Layering of stomach contents in drowning cases in post-mortem computed tomography compared to forensic autopsy.

Background: In forensic autopsy, the analysis of stomach contents is important when investigating drowning cases. Three-layering of stomach contents may be interpreted as a diagnostic hint to drowning due to swallowing of larger amounts of water or other drowning media. The authors experienced frequent discrepancies of numbers of stomach content layering in drowning cases between post-mortem computed tomography (PMCT) and autopsy in forensic casework. Therefore, the goal of this study was to compare layering of stomach contents in drowning cases between PMCT and forensic autopsy.

Methods: Drowning cases (n = 55; 40 male, 15 female, mean age 45.3 years; mean amount of stomach content 223 ml) that received PMCT prior to forensic autopsy were retrospectively analyzed by a forensic pathologist and a radiologist. Number of layers of stomach content in PMCT were compared to number of layers at forensic autopsy.

Results: In 28 of the 55 evaluated drowning cases, a discrepancy between layering of stomach contents at autopsy compared to PMCT was observed: 1 layer at autopsy (n = 28): 50% discrepancy to PMCT, 2 layers (n = 20): 45% discrepancy, and 3 layers (n = 7): 71.4% discrepancy. Sensitivity of correctly determining layering (as observed at forensic autopsy) in PMCT was 52% (positive predictive value 44.8%). Specificity was 46.6% (negative predictive value 53.8%). In a control group (n = 35) of non-drowning cases, three-layering of stomach contents was not observed.

Conclusion: Discrepancies of observed numbers of stomach content layers between PMCT and forensic autopsy are a frequent finding possibly due to stomach content sampling technique at autopsy and movement of the corpse prior to PMCT and autopsy. Three-layering in PMCT, if indeed present, may be interpreted as a hint to drowning.