摘要
Introduction: A novel three dimensional approach for aneuploidy screening in the first trimester of pregnancy was developed in which risk assessment derives directly from comparing the plotted data of nuchal translucency, pregnancy associated plasma protein A (PAPP-A), and free β-human chorionic gonadotropin (fβ-hCG) of an examined fetus with similar coordinates of fetuses with already known health status. Under this approach, it is possible to utilize either a ‘box’ or a ‘sphere’ model. In either case, optimal volume sizes and the benefits of adopting a ‘minimum number of required fetuses’ (MNR) have not yet been investigated;and for the box model, two modifications, called ‘empty box results positive’ (EB+) and ‘simulation’ (SIM), provide additional options. It was the aim of this study to analyze which of the two models and their variants provides the best test performance. Methods: The study cohort was divided into a reference collective (n = 10,954) and a test collective (n = 4239). The test collective was examined repeatedly, with another model and modification used on each occasion. Test performances were compared by the area under curve (AUC) of receiver operating characteristics (ROC) curves. Results: The sphere model was inferior to the box model when optimal volumes were used with the latter and combined with the modifications EB+ and Sim. EB+ increased the number of assessable fetuses while Sim improved the test performance. MNR improved neither the box nor the sphere model. Conclusion: A new, optimized model in line with the obtained results should be developed and tested in further studies.
Introduction: A novel three dimensional approach for aneuploidy screening in the first trimester of pregnancy was developed in which risk assessment derives directly from comparing the plotted data of nuchal translucency, pregnancy associated plasma protein A (PAPP-A), and free β-human chorionic gonadotropin (fβ-hCG) of an examined fetus with similar coordinates of fetuses with already known health status. Under this approach, it is possible to utilize either a ‘box’ or a ‘sphere’ model. In either case, optimal volume sizes and the benefits of adopting a ‘minimum number of required fetuses’ (MNR) have not yet been investigated;and for the box model, two modifications, called ‘empty box results positive’ (EB+) and ‘simulation’ (SIM), provide additional options. It was the aim of this study to analyze which of the two models and their variants provides the best test performance. Methods: The study cohort was divided into a reference collective (n = 10,954) and a test collective (n = 4239). The test collective was examined repeatedly, with another model and modification used on each occasion. Test performances were compared by the area under curve (AUC) of receiver operating characteristics (ROC) curves. Results: The sphere model was inferior to the box model when optimal volumes were used with the latter and combined with the modifications EB+ and Sim. EB+ increased the number of assessable fetuses while Sim improved the test performance. MNR improved neither the box nor the sphere model. Conclusion: A new, optimized model in line with the obtained results should be developed and tested in further studies.