15283-B-Blokker

173 General discussion 9 Pathologies related to air/ gas MIA outperforms CA in pathology where air/ gas is involved, especially in cases of (massive) air embolus and pneumothorax. 57,129, chapter 5 Conversely, the lungs are difficult to examine due to livor mortis in the dependent parts of the lungs. To improve the diagnostic accuracy for small pulmonary lesions, CT scans may be combined with mechanical ventilation, which inflates the lungs to expand their volume and reduce the lung density. 121-123 In our cohort, however, we did not miss any small lesions in the lungs. MIA did falsely diagnose one acute pneumonia, and one aspergillus infection in the lungs was missed upon histological examination. Collections of air may also be present in the case of intestinal pathologies. On imaging, bubbles of gas within the intestinal wall (pneumatosis) and air collections around the portal vein are a sign of ischemia. 103,111,210 They may, however, also represent the beginning of post-mortem putrefaction. To differentiate the two, it may be helpful to compare the post-mortem scans to those made just before death, if available. In this cohort, we also had difficulties in identifying or confirming ischemic colitis as the cause of death with MIA, and some intestinal pathologies were difficult to distinguish: fungi in the small intestine versus an infectious enteritis, and radiation mucositis versus typhlitis. Cerebral pathologies The main issue regarding neurological diseases was the fact that consent fromnext-of- kinwas required to obtain brain biopsies for histological examination. MIA brain biopsies were permitted in 24 cases and conventional brain autopsy in 38 cases, among these there were only 20 cases with consent for both MIA brain biopsies and conventional brain autopsy. However, with or without consent for brain biopsy or dissection, MIA always included specific imaging of the brain, which was better than no examination at all. Generally, large infarctions and haemorrhages are clearly visible at imaging, but small subdural haemorrhages may be hard to detect, for they may be pushed away by the post-mortem swelling of the brain. In our cohort were several cases, in which the imaging findings of the brain alone could not establish certain diagnoses, while these diagnoses could not be confirmed by brain biopsy or brain autopsy. Due to the lack of histological correlation, the accuracy of MIA for brain pathology cannot be exactly determined in this cohort. Nonetheless, in 11% of the cases cerebral pathology, often based on radiological evaluation only, was related to the cause of death. It was thereby the fourth most common cause of death in our cohort, emphasizing the importance of post-mortem brain examination.