Magnetic resonance imaging (MRI) was introduced more than a decade ago as a noninvasive alternative to conventional invasive classic autopsy. Several studies have reported on the diagnostic usefulness of postmortem virtuopsy. No studies have investigated the acceptance of both autopsy and virtuopsy, factors influencing the decision to accept or decline either technique, or the reliability of postmortem virtuopsy compared with conventional autopsy.
This multicenter prospective study investigated the maternal acceptance of virtuopsy, including postmortem MRI and computed tomography, compared with classic invasive autopsy, and also evaluated the reliability and confidence of this noninvasive method to diagnose the normality/abnormality of various fetal anatomical structures. Subjects were 96 mothers with 102 fetuses/neonates following termination of pregnancy (TOP), intrauterine fetal death (IUFD), or neonatal death. All mothers were given a choice of virtuopsy or conventional autopsy, both or neither. Multivariable logistic regression analysis was used to assess the possible influence of the following variables on maternal acceptance of virtuopsy and/or conventional autopsy: age of the mother, gestational age at TOP or delivery after IUFD, order of pregnancy, parity, religion, type of caregiver obtaining consent, and reason for death. When parents consented to both autopsy and virtuopsy for fetuses at 20 or more weeks of gestation or neonates, the confidence that virtuopsy had correctly diagnosed normality/abnormality of various anatomical structures was determined applying a scale using conventional autopsy as the criterion standard. The scale values ranged from 0 (definitely abnormal) to 100 (definitely normal). The confidence of a correct diagnosis in the fetuses/neonates for every anatomical structure was analyzed separately in each group at autopsy or virtuopsy.
Among the 96 women, 99% (95/96) consented to virtuopsy, 61.5% (59/96) to both conventional autopsy and virtuopsy, and the remaining 37.5% (36/96) to virtuopsy alone. Acceptance by the mother of conventional autopsy was independently positively related to singleton pregnancy, non-Moslem mother, earlier gestation at TOP or delivery following IUFD, and a maternal-fetal medicine specialist obtaining the consent.
Among the fetuses/neonates examined with both autopsy and virtuopsy, 33 were 20 weeks or older; all 33 had invasive autopsy, and 19 of these had a full autopsy including the brain. In fetuses with normal anatomical structures at autopsy, there were high confidence scores (>80) with virtuopsy for the brain, skeleton, thoracic organs except the heart and lung lobations, and abdominal organs except the pancreas, ureters, bladder, genitals, duodenum, and bowel. For fetuses with abnormal anatomical structures at autopsy, virtuopsy had high confidence scores (<20) for detection of anomalies in these same anatomical structures. Virtuopsy diagnosed 3 brain anomalies in addition to those observed at conventional autopsy.
These findings demonstrate that magnetic resonance virtuopsy is accepted by nearly all mothers, whereas conventional autopsy is accepted by about 60% of mothers. Refusal depends mainly on factors that cannot be controlled, such as gestational age at TOP or religion.
Department of Radiology, University Hospital Brugmann, Brussels; Department of Radiology, University Hospital Gasthuisberg, Leuven; Department of Radiology, University of Brussels, VUB, Department of Obstetrics and Gynaecology, University Hospital Brugmann, and Department of Obstetrics, University of Brussels, VUB, Brussels; Department of Pathology, University Hospital Gasthuisberg, Leuven; Department of Pathology, University Hospital Brugmann, Brussels; and Department of Obstetrics and Gynaecology, University Hospital Gasthuisberg, Leuven, Belgium.