15283-B-Blokker

204 methods. None of the alternative methods performed as well as CA, but the highest sensitivity for cause of death was reasonably good: 90.9% (two studies combined, suspected duplicates excluded). This was achieved with a combination of CT, CT- angiography and biopsies. An important limitation of our review was, apart from the few studies with a blinded comparison and many studies with only few cases, that the data presented in the sixteen studies was too heterogeneous for meaningful meta- analysis of outcomes. We decided that further investigation was required in order to establish a feasible alternative to CA in clinical practice, preferably in a larger study group. In chapter 5 , we present our own prospective, single center, blinded cross-sectional study that compares the diagnostic performance of a minimally invasive autopsy (MIA) and CA. For this, 99 deceased adult patients were included, for whom next-of-kin gave consent to perform both MIA and CA. Our MIA comprised unenhanced whole-body CT and MRI scans, and image-guided biopsies of vital organs and pathological lesions detected on imaging. MIA and CA agreed on cause of death in 91 cases; MIA established the right cause of death in 5 and CA in 3 of the remaining cases. Of the 288 diagnoses directly related to the cause of death, MIA diagnosed 259 and CA 224; 200 diagnoses were found by both methods. Finally, MIA and CA performed equally in answering the specific questions of clinicians. The method we applied for MIA, which is feasible in any larger general hospital and generates a permanent and auditable record, appeared to be a valid alternative for CA. Importantly, no less than 17% of the causes of death and 43% of the 288 diagnoses related to cause of death were clinically unsuspected, emphasizing the necessity of performing post-mortem examination for the purpose of healthcare quality control. The study in chapter 6 is a sub-analysis of this same cohort, analyzing the diagnostic accuracy of MIA for the detection of ischemic heart disease, which forms a major burden of disease in Western countries. In this analysis, where CA served as the gold standard, the combination of specialized MRI scans and targeted heart-biopsies reached a sensitivity of 97% for the detection of acute ischemia and a specificity of 95%. However, without the heart-biopsies the MRI scans had low sensitivity for both acute ischemia and chronic ischemia, emphasizing the value of histological examination in the post-mortem. Whereas other studies stress the need of post-mortem angiography for establishing a cardiac cause of death, we find that our MRI scans combined with targeted heart-biopsies suffice. Through the feasibility study described in chapter 7 , we found that post-mortem tissue biopsies taken at MIA yield a sufficient amount of RNA, which is also of sufficient quality, especially if the biopsies are obtained shortly after death. With such good RNA quality, these biopsies are useful for molecular translational research, including gene Summary