Abstract
Regular audit of results of prenatal screening for congenital heart disease (CHD) is crucial to ensure reliable prenatal diagnosis. We aimed to assess the accuracy of prenatal diagnosis of major CHD between 1996 and 2013. During the study period, prenatal detection of major CHD improved from 4.5% to 71.0% (p<.001). Prenatal diagnoses on 628 live born children and terminated pregnancies were compared with postnatal findings or autopsy reports. The proportion of correct diagnoses increased throughout the study period from 42.9% in 1996 and reached 88.2% in 2013 (p<.001). A total of 32 foetuses with suspected major CHD were terminated though no major CHD was found at autopsy. In these pregnancies, termination was mainly performed due to other anomalies in the foetus.
Along with improved detection of major CHD, the validity of a prenatal diagnosis is increasing. No cases of misinterpreted major CHD resulted in the termination of a healthy foetus in this study.
What is already known on this subject? Prenatal diagnosis of isolated congenital heart disease (CHD) correlates well with lesions found during autopsy performed in terminated foetuses. Few studies have assessed the accuracy of prenatal diagnosis of major CHD in live born children, cases with associated anomalies and the time trend in validity.
What the results of this study add? This study illustrates that the validity of prenatal diagnosis of major CHD is increasing. Prenatal diagnoses in terminated pregnancies as well as in live born children is high except for coarctation of the aorta and atrioventricular septal defects. Chromosomal anomalies are associated with lower accuracy of prenatal diagnosis.
What the implications are of these findings for clinical practice and/or further research? Prenatal diagnosis is an accurate tool for detecting major CHD. Misinterpretation has not led to the termination of a healthy foetus; however, this study illustrates that vigilant care should be placed on the cardiac evaluation when termination is considered due to the cardiac defect.
Impact statement
Acknowledgements
We would like to thank the following for their aid in the collection of data: Eva Andersen, Department of Obstetrics and Gynaecology, Randers Hospital; Louise Thomsen Schmidt Arenholt, Department of Obstetrics and Gynaecology, Hjørring Hospital; Lise Bjerglund, Department of Paediatrics, Nykøbing Falster Hospital; Susanne Ebbe, Department of Paediatrics, Viborg Hospital; Annie Ellermann, Department of Paediatrics, Holbaek Hospital; Joan Frederiksen, Department of Paediatrics, Randers Hospital; Marta Gavai, Department of Obstetrics and Gynaecology, Thisted Hospital; Birgitte Johnsen, Department of Obstetrics and Gynaecology, Esbjerg Hospital; Lone Krebs, Department of Obstetrics and Gynaecology, Holbaek Hospital; Bjarne Kristensen, Department of Obstetrics and Gynaecology, Odense University Hospital; Esther Kristensen, Department of Paediatrics, Esbjerg Hospital; Finn Lauszus, Department of Obstetrics and Gynaecology, Herning Hospital; Annemette Lykkebo, Department of Obstetrics and Gynaecology, Kolding Hospital; Katrin Löser, Familiecenteret, Sygehus Sønderjylland; Jonas Manthorpe, Department of Paediatrics, Zealand University Hospital; Eva Mosfeldt, Department of Paediatrics, Herlev Hospital; Dorthe Nielsen, Department of Paediatrics, Kolding Hospital; Henrik Nissen, Department of Cardiology, Odense University Hospital; Lia Pedersen, Department of Paediatrics, Aalborg Hospital; Heidi Petersen, Department of Paediatrics, Naestved Hospital; Heidi Sharif, Department of Obstetrics and Gynaecology, Naestved Hospital; Charlotte Søndergaard, Department of Paediatrics, Herning Hospital; Dorthe Vestergaard, Department of Paediatrics, Nordsjaellands Hospital; Tina Vinther-Madsen, Department of Obstetrics and Gynaecology, Zealand University Hospital; Helle Zingenberg, Department of Obstetrics and Gynaecology, Herlev Hospital; Karen Øberg, Department of Obstetrics and Gynaecology, Nordsjaellands Hospital.
Data availability statement
The data that support the findings of this study are openly available in Mendeley at DOI:10.17632/wvxtsmmdkd.1.