VP25.01: The correlation between ultrasonographic diagnosis of fetal brain malformations and postmortem findings: a retrospective study

The aim of this study was to correlate fetal brain anomalies detected by ultrasound examination with those identified at autopsy following termination of pregnancy (TOP); and to develop a categorisation method based on the level of agreement between the two techniques. A retrospective case series on 104 TOP performed over a 7-year period, from 2013 to 2019. Prenatal neurosonographic examinations were performed at the Fetal Neurology Clinic. Autopsies were performed in collaboration between TLVMC and Meir Medical Centre Pathology Departments. Five levels of agreement were defined to describe the degree of correlation in diagnosis between the two. In 80% (83/104) of the cases the correlation was considered as “optimal” with or without additional findings observed at the autopsy. Only a “partial” correlation, with or without additional information was observed in 12.5% (13) of cases. In 7.7% (8) of cases, observations made by ultrasound were not confirmed at autopsy; in three (2.9%) an alternative diagnosis was proposed and in five (4.8%) the autopsy was considered uninformative. Diagnostic accuracy averege rates for all ultrasonographic diagnoses were sensetivity of 88.9% (42.5%, 98.5%), specificity of 94.3% (78.1%, 97%), PPV of 91% (73%, 94.2%) and NPV of 87.2% (85.2%, 93.1%). Our study demonstrates an excellent correlation between dedicated neurosonography and autopsy findings in the field of fetal brain malformations. Despite this high correlation in a considerable number of cases the autopsy adds important information not demonstrable by imaging. The risk of false-positive US diagnoses although low should always be considered.