Assessment of fetal heart aortic and pulmonary valve annulus area by three-dimensional ultrasonography: reference curves and applicability in congenital heart diseases.

OBJECTIVES

To determine reference curves for fetal aortic and pulmonary valve annulus area by three-dimensional ultrasonography using the spatio-temporal image correlation (STIC) in the rendering mode, and to ascertain its applicability in congenital heart disease (CHD).

METHODS

We performed a retrospective cross-sectional study of 328 normal fetuses and 42 fetuses with CHD between 20 and 33 weeks 6 days of gestation. The outflow plane view of the great vessels was used to measure the areas of the valvar annuli, and the measurements were performed in systole. A linear regression model adjusted according to the determination coefficient (R2) was utilized to construct the reference intervals. The concordance correlation coefficient (CCC) was used to calculate the reproducibility of the mitral and tricuspid valve areas.

RESULTS

The mean ± standard deviation (SD) of the aortic and pulmonary valve annulus areas ranged from 6.6 ± 1.2 to 32.9 ± 1.1 mm2 and 10.7 ± 1.3 to 40.3 ± 1.2 mm2, respectively. We observed a linear relationship and strong positive correlation between the area of the aortic and pulmonary valve annuli with r=0.97 and 0.96, respectively. Good intra (CCC=0.99) and interobserver agreement (CCC=0.98) was observed for the measurement of the aortic valve annulus area. A good intra (CCC=0.99) and interobserver (CCC=0.97) agreement was also observed for the measurement of the pulmonary valve annulus area. The mean ± SD of the difference of the areas of the aortic and pulmonary valve annuli between the normal fetuses and those with CHD were -1.801 ± 1.429 mm2 (p=0.208) and -1.033 ± 1.467 mm2 (p<0.0001), respectively.

CONCLUSIONS

The reference curves for the areas of the aortic and pulmonary valve annuli of fetal hearts were determined, and showed good inter and intraobserver reproducibility. The constructed reference curves showed applicability in different types of CHD.