15283-B-Blokker

80 Chapter 5 ABSTRACT Background: The worldwide decline of clinical conventional autopsies (CA), among others due to its invasiveness, means a loss for healthcare quality control, calling for development of alternative methods. Objective of this study was to compare diagnostic performance of a minimally invasive autopsy (MIA) and CA. Methods: Prospective, single center, cross-sectional study in an academic hospital. The MIA procedure, combining MRI, CT and image-guided biopsies of main organs and pathological lesions detected on imaging, was followed by CA on the same case. MIA and CA were performed by separate teams, blinded to each other. Primary outcome measure: performance of MIA and CA in establishing cause of death. Secondary outcome measures: diagnostic yield of MIA and CA for all, major and grouped major diagnoses, frequency of clinically unsuspected findings, and percentage of answered clinical questions. The different setting of MIA (research) and CA (routine) precluded the use of CA as gold standard. Results: Out of 2197 adult patients, who died in 2012 through 2014, 295 underwent CA; for 139 cases there was also consent for MIA; 99 could be included in the study cohort. Agreement between MIA and CA for cause of death existed in 91/99 cases, agreement with consensus cause of death in 96/99 for MIA, and 94/99 for CA (P=0.727). 288 grouped major diagnoses were directly related to consensus cause of death, MIA diagnosed 259 (90%) and CA 224 (78%), 200 (69.4%) were found by bothMIA and CA. 17% of cause of death, and 43% of grouped major diagnoses were clinically not suspected. MIA answered 189 (86·3%) and CA 182 (83·1%) of 219 clinical questions. Conclusions: MIA and CA performed equally well in detecting cause of death. MIA resulted in a higher yield of diagnoses. MIA appears a valid alternative for CA, resulting in a permanent, auditable record of the deceased. The frequency of clinically unsuspected findings underscores the lasting importance of post-mortem investigation.