Faculty Bibliography

OBJECTIVE: To develop and evaluate a method of estimating patient-specific risk for fetal loss by combining maternal characteristics with serum markers. STUDY DESIGN: Data were obtained on 36,014 women from the FaSTER trial. Separate likelihood ratios were estimated for significant maternal characteristics and serum markers. Patient-specific risk was calculated by multiplying the incidence of fetal loss by the likelihood ratios for each maternal characteristic and for different serum marker combinations. RESULTS: Three hundred eighteen women had fetal loss < 24 weeks (early) and 103 > 24 weeks (late). Clinical characteristics evaluated included maternal age, body mass index, race, parity, threatened abortion, previous preterm delivery, and previous early loss. Serum markers studied as possible predictors of early loss included first-trimester pregnancy-associated plasma protein A and second-trimester alpha-fetoprotein, and unconjugated estriol. A risk assessment for early loss based on all of these factors yielded a 46% detection rate, for a fixed 10% false-positive rate, 39% for 5% and 28% for 1%. The only significant marker for late loss was inhibin A. The detection rate was 27% for a fixed 10% false-positive rate and only increased slightly when clinical characteristics were added to the model. CONCLUSION: Patient-specific risk assessment for early fetal loss using serum markers, with or without maternal characteristics, has a moderately high detection. Patient-specific risk assessment for late fetal loss has low detection rates.

Twin reversed arterial perfusion (TRAP) sequence, also known as acardiac malformation, is a rare complication unique to monozygotic multiple gestations. It occurs in approximately 1 per 35,000 pregnancies and in 0.3% of all monozygotic twin gestations. The malformation is characterized by the lack of a well-defined cardiac structure in one twin (acardiac), which acts as a parasite that is hemodynamically dependent on the normal (pump) twin. The malformation in the acardiac twin is always incompatible with survival regardless of its extent, whereas the prognosis of the pump twin may vary considerably. Common complications that affect the prognosis of the pump twin include complications of prematurity secondary to preterm delivery and polyhydramnios as well as congestive heart failure due to the expended cardiac demand. Without prompt detection, follow-up, and treatment, mortality rates for pump twins have been noted to be as high as 50% to 70%. Early sonographic diagnosis, appropriate follow-up, and intervention via cord occlusion when indicated are the mainstays of prenatal care in cases of TRAP sequence and have been associated with substantial improvements in survival. We report a case that emphasizes the added value of 3-dimensional (3D) and color Doppler techniques for both diagnosis of TRAP sequence and detection of poor prognostic features in the first or early second trimester

The widespread use of sonography as a screening tool for fetal anomalies has facilitated prenatal detection of several fetal conditions characterized by urinary tract dilatation. These conditions are more common in male fetuses and are generally a result of an anatomic defect causing obstruction along the urinary tract system. Although the prognosis of these conditions largely depends on the specific anomaly, several poor prognostic factors have been described. These factors include detection at an early gestational age, bilateral marked dilatation, a persistently obstructed bladder, oligohydramnios causing pulmonary hypoplasia, and the presence of associated fetal or chromosomal anomalies. We report a case in which a male fetus at 14 weeks' gestation had a diagnosis of rapidly progressing bilateral hydronephrosis, massive bladder dilatation, hydroureter, and a surprisingly normal amniotic fluid volume. Serial sonographic surveillance assisted us in obtaining the correct diagnosis, which was important for adequately consulting the patient regarding the fetal prognosis in the affected index pregnancy as well as the likelihood of recurrence in future gestations

OBJECTIVE: To demonstrate that individualized optimal fetal growth norms, accounting for physiologic and pathologic determinants of fetal growth, better identify normal and abnormal outcomes of pregnancy than existing methods. METHODS: In a prospective cohort of 38,033 singleton pregnancies, we identified 9,818 women with a completely normal outcome of pregnancy and characterized the physiologic factors affecting birth weight using multivariable regression. We used those physiologic factors to individually predict optimal growth trajectory and its variation, growth potential, for each fetus in the entire cohort. By comparing actual birth weight with growth potential, population, ultrasound, and customized norms, we calculated for each fetus achieved percentiles, by each norm. We then compared proportions of pregnancies classified as normally grown, between 10th and 90th percentile, or aberrantly grown, outside this interval, by growth potential and traditional norms, in 14,229 complicated pregnancies, 1,518 pregnancies with diabetes or hypertensive disorders, and 1,347 pregnancies with neonatal complications. RESULTS: Nineteen physiologic factors, associated with maternal characteristics and early placental function, were identified. Growth potential norms correctly classified significantly more pregnancies than population, ultrasound, or customized norms in complicated pregnancies (26.4% compared with 18.3%, 18.7%, 22.8%, respectively, all P<.05), pregnancies with diabetes or hypertensive disorders (37.3% compared with 23.0%, 28.0%, 34.0%, respectively, all P<.05) and neonatal complications (33.3% compared with 19.7%, 24.9%, 29.8%, respectively, all P<.05). CONCLUSION: Growth potential norms based on the physiologic determinants of birth weight are a better discriminator of aberrations of fetal growth than traditional norms. LEVEL OF EVIDENCE: II

OBJECTIVE: The purpose of this study was to evaluate the usefulness of virtual spherical tissue sampling using 3-dimensional (3D) ultrasound power Doppler angiography to enhance differentiation between normal and pathologic ovaries. METHODS: Twenty-seven cases with ovarian tumors were analyzed: 14 with invasive cancers and 13 with borderline tumors confirmed by surgery. The control subjects consisted of 53 healthy ovulating women. Ultrasound scans were done, and 3D volumes were analyzed with 3-/4-dimensional software for personal computers based on 3D vascularity indices: the vascularization index, flow index, and vascularization-flow index. A virtual spherical tissue sample of 1 cm(3) was taken from the place of the highest vessel density contained completely within the contours of the ovary. Calculations for the whole solid volume were done for comparison. RESULTS: Vascularity indices for both 1-cm(3) spherical samples and whole dense parts of the ovaries were compared in the following groups: (1) ovarian tumors versus controls, (2) normal ovaries in the proliferative versus secretory phase, (3) invasive cancers versus borderline tumors, (4) invasive cancers versus normal ovaries, and (5) borderline tumors versus normal ovaries. Spherical 1-cm(3) sampling achieved a higher degree of discrimination between the groups compared with the whole solid-part approach. CONCLUSIONS: Spherical 1-cm(3) sampling of ovarian tissue with 3D ultrasound power Doppler angiography is a sensitive and promising approach to differentiate between ovarian tumors and normal ovaries. It opens the possibility to implement objective computerized positioning, standardized comparison, and analysis of ovarian tumors

OBJECTIVE: Comparison of contingent, step-wise and integrated screening policies. METHODS: Mid-trimester Down syndrome risks were retrospectively calculated from FaSTER trial data. For contingent screening, initial risk was calculated from ultrasound measurement of nuchal translucency (NT), maternal serum pregnancy-associated plasma protein (PAPP)-A and free beta-human chorionic gonadotrophin (hCG) at 11-13 weeks, and classified positive (>1 in 30), borderline (1 in 30-1500) or negative. Borderline risks were recalculated using alpha-fetoprotein, hCG, unconjugated estriol (uE3) and inhibin at 15-18 weeks, and reclassified as positive (>1 in 270) or negative. For step-wise screening, initial negative risks were also recalculated. For integrated screening, a single risk was calculated from NT, PAPP-A and the second trimester markers. RESULTS: There were 86 Down syndrome and 32,269 unaffected pregancies. The detection rate for contingent screening was 91% and false-positive rate was 4.5%; initial detection rate was 60%, initial false-positive rate was 1.2% and borderline risk was 23%. Step-wise screening had 92% detection rate and 5.1% false-positive rate; integrated screening had 88% and 4.9% respectively. CONCLUSION: As predicted by modelling, the contingent screening detection rate for a fixed false-positive rate is comparable with step-wise and integrated screening, but substantially reduces the number needing to return for second trimester testing