Congenital anomalies are the second leading cause of perinatal mortality in the United States after premature birth. Advances in imaging techniques such as ultrasound have allowed the detection of many anatomical defects before birth. The goal of this project is to improve the diagnosis and treatment of fetal disease and congenital anomalies. The Perinatology Research Branch has initiated a project to use three-dimensional ultrasound to delineate normal fetal anatomy and to improve the detection of congenital anomalies and the estimation of fetal weight and growth. The diagnostic power of this technology is illustrated by the improved detection of cleft lip, vasa previa, and congenital anomalies. This year, the branch reported a method to estimate fetal weight by using fractional thigh volume with three-dimensional ultrasound. We also described a methodology to standardize iliac angle measurements in second trimester fetuses and conducted a study to determine whether "genetic ultrasound" is a cost-effective approach to the detection of trisomy 21.

Birthweight Prediction by Three-Dimensional Ultrasonography: Fractional Limb Volume
Romero
Estimation of fetal weight is critical for the diagnosis of fetal growth disorders (intrauterine growth retardation and macrosomia). The standard approach for estimation of fetal size and weight is the use of two-dimensional ultrasound. However, sonographic fetal biometry can generate errors as great as 15 percent of the actual birth weight, and many formulas used for this purpose tend to overestimate weights in smaller fetuses and underestimate weights in larger fetuses. The branch introduced the concept of fractional limb volume, a new sonographic parameter obtained with the use of three-dimensional ultrasound for the estimation of fetal weight. At the same time, we validated the estimate's reliability and developed new birthweight prediction models. A model to predict fetal weight by using a combination of the abdominal circumference and the fractional thigh volume resulted in estimations of fetal weight that had a 0.5 percent systematic error and a 7 percent random error, which compare favorably to the widely used model based on conventional two-dimensional ultrasound (9 percent systematic error and 9 percent random error).

Fetal Iliac Angle Measurements by Three-Dimensional Ultrasonography
Romero
Dysmorphic pelvic features are common among newborn infants and adults with trisomy 21. Bilateral widening and flaring of the iliac wings have been previously reported in newborns affected by trisomy 21, and a previous study using an iliac angle threshold of 90 degrees was found to have a sensitivity of 37 percent and a false positive rate of 12.8 percent in the detection of trisomy 21. However, fetal iliac angles are difficult to determine reliably by using two-dimensional ultrasonography (intra- and inter-observer correlation coefficients of 0.7 and 0.62, respectively). Three-dimensional multiplanar views of the fetal pelvis were used to standardize iliac angle measurements from normal second-trimester fetuses as well as a group affected by trisomy 21. The mean axial angle for normal fetuses was significantly less than observed in fetuses with trisomy 21. Iliac angles did not correlate with gestational age. Angle measurements were reproducible when obtained in an axial plane while those derived by coronal views were not. An axial angle of 87 degrees correctly identified 56 percent of fetuses with trisomy 21 with a false positive rate of 5 percent (79 degrees was the value obtained for normal fetuses). This information can be incorporated into the assessment of risk for trisomy 21.

Combined Use of Genetic Sonography and Maternal Serum Triple-Marker Screening as an Effective Method for Increasing the Detection of Trisomy 21 in Women Younger than 35 Years
DeVore, Romero
The current standard of practice for the detection of trisomy 21 in women less than 35 years of age is to conduct triple-marker screening (maternal serum alpha-fetoprotein, estriol, and human chorionic gonodotropin) followed by amniocentesis in patients who are defined as high risk or having a risk of trisomy 21 of 1:10 to 1:190. Non-high-risk patients are not offered further diagnostic testing. The branch conducted a study to determine whether genetic ultrasound of fetuses considered to be at moderate risk (1:190 to 1:1,000) following maternal triple-marker screening increases the detection of trisomy 21, whether the method is cost-effective, and whether it reduces the number of amniocenteses required to detect a single fetus with trisomy 21. The term "genetic ultrasound" refers to a comprehensive examination of fetal external and internal anatomy with sonography to identify biometric and anatomical markers that, when detected, increase the risk for trisomy 21. Modeling of the sensitivity of genetic ultrasound and the false-positive rate indicated that offering genetic sonography to women less than 35 years of age at moderate risk would increase the detection rate of trisomy 21. In addition, it is cost-effective, and would reduce the number of amniocenteses required to detect a fetus affected affected by this condition.