Faculty Bibliography

Abstract Objective: To compare pregnancy outcomes in twin pregnancies based on maternal pre-pregnancy body mass index (BMI). Methods: Historical cohort study of all twin pregnancies >24 weeks managed by one maternal-fetal medicine practice from 2005 to 2012. We compared pregnancy outcomes between pre-pregnancy obese (BMI >/=30 kg/m(2)) and normal weight women (BMI 18.5-24.99 kg/m(2)). We also compared pre-pregnancy normal weight women to overweight women (BMI 25-29.99 kg/m(2)) and underweight women (BMI <18.5 kg/m(2)). Chi square, Fisher's exact test, Student's t-test, and one-way ANOVA were used as appropriate. A p value of <0.05 was considered significant. Results: Five hundred fourteen patients with twin pregnancies were included. Pre-pregnancy obesity was associated with gestational hypertension (34.1% versus 17.9%, p = 0.011), preeclampsia (27.3% versus 14.4%, p = 0.028), and gestational diabetes (22.2% versus 4.7%, p < 0.001). Pre-pregnancy overweight was associated with gestational diabetes (13.7% versus 4.7%, p = 0.002). Pre-pregnancy underweight was not associated with any adverse pregnancy outcomes. Comparing outcomes across normal weight, overweight, and obese women, the rates of gestational diabetes and gestational hypertension increased significantly across the three groups. Conclusion: In patients with twin pregnancy, pre-pregnancy obesity is associated with adverse pregnancy outcomes, including gestational diabetes, gestational hypertension, and preeclampsia.

Objective Twin pregnancy is associated with an increased incidence of preeclampsia. However, it is unknown if the risk factors for preeclampsia in twin pregnancies are the same as those in singleton pregnancies. Methods Case-control analysis of all twin pregnancies managed by one maternal-fetal medicine practice from 2005 to 2012. Patients with chronic hypertension were excluded, as were monochorionic-monoamniotic twins. We compared patient and pregnancy characteristics between patients who did and did not develop preeclampsia, according to standard American College of Obstetricians and Gynecologists definitions. Odds ratios, adjusted odds ratios (aORs), and 95% confidence intervals (CIs) were obtained using chi-square analysis and logistic regression. Results Of the patients with twin pregnancies, 513 were included, and 76 (14.8%) patients developed preeclampsia. On univariable analysis, the risk factors associated with preeclampsia in twin pregnancies were egg donation, nonwhite race, nulliparity, prepregnancy obesity, and gestational diabetes. On adjusted analysis, the risk factors independently associated with preeclampsia were egg donation (aOR 2.409, 95% CI 1.051, 5.524) and prepregnancy obesity (aOR 2.367, 95% CI 1.079, 5.192). Conclusions In twin pregnancy, the risk factors independently associated with preeclampsia are egg donation and prepregnancy obesity.

OBJECTIVES: The objective of this article is to determine whether there were differences in first trimester serum analytes between cases of placenta previa with and without accreta. METHODS: Cases of placenta previa in which the patient had first trimester aneuploidy screening were identified. Pregnancy-associated plasma protein A (PAPP-A) and free beta human chorionic gonadotropin (fbhCG) MoMs were compared with those with an accreta. Accreta cases were also compared with published distributions to determine significance and to develop likelihood ratios based on MoM values. RESULTS: Eighty-two cases of previa were identified, including 16 with a histological diagnosis of placenta accreta. The median PAPP-A MoM of 1.68 in accreta was significantly greater than that of 0.98 in non-accreta (P = 0.002). For fbhCG, the median MoM was 1.00 and 1.01 in accreta and non-accreta, respectively. Of the 16 patients with accreta, 14 (87.5%, 95% confidence interval: [61.6%, 98.4%]) had PAPP-A MoM above 1.0. Six of 16 (37.5%) accreta cases were above the 90th percentile of the unaffected distribution. The likelihood ratios for accreta were 0.5, 2.0, and 3.0. PAPP-A MoMs were 0.19, 2.11, and 4.27, respectively. CONCLUSIONS: First trimester PAPP-A levels may be useful in identifying pregnancies at high risk for placenta accreta. Larger studies could incorporate both clinical and biochemical data into a risk algorithm.

Objectives- The purpose of this study was to estimate the prevalence and persistence rate of vasa previa in at-risk pregnancies using a standardized screening protocol. Methods- We conducted a descriptive study of patients with a diagnosis of vasa previa from a single ultrasound unit between June 2005 and June 2012. Vasa previa was defined as a fetal vessel within 2 cm of the internal cervical os on transvaginal sonography. Screening for vasa previa using transvaginal sonography with color flow mapping was performed routinely in the following situations: resolved placenta previa, prior pregnancy with vasa previa, velamentous insertion of the cord in the lower uterine segment, placenta succenturiata in the lower uterine segment, and twin gestations. Results- A total of 27,573 patients were referred to our unit for fetal anatomic surveys over the study period. Thirty-one cases of vasa previa were identified, for an incidence of 1.1 per 1000 pregnancies. Twenty-nine cases had full records available for analysis. Five patients (17.2%) had migration and resolution of the vasa previa. When the diagnosis was made during the second trimester (<26 weeks), there was a 23.8% resolution rate (5 of 21); when the diagnosis was made in the third trimester, none resolved (0 of 8 cases). Of the 24 pregnancies (5 twin gestations and 19 singleton gestations) with persistent vasa previa, there was 100% perinatal survival and a median length of gestation of 35 weeks (range, 27 weeks 5 days-36 weeks 5 days). No known missed cases were identified over the study period. Conclusions- The use of standardized screening for vasa previa based on focused criteria was found to be effective in diagnosing vasa previa, with a 100% survival rate. Vasa previa diagnosed during the second trimester resolves in approximately 25% of cases.

OBJECTIVE: To describe pregnancy characteristics associated with the occurrence of spontaneous version in twin pregnancies from 20 weeks until delivery. STUDY DESIGN: Review of ultrasound and pregnancy data for all twin pregnancies delivered >24 weeks by one Maternal-Fetal Medicine practice from June 2005-May 2012. For each 4-week gestational age window, fetal presentations were recorded, as well as the likelihood of the final presentation being vertex for Twin A and vertex for both twins. Case-control analysis was performed to estimate associations between pregnancy characteristics and spontaneous version of Twin A. RESULTS: A total of 491 twin pregnancies were included. The distribution of fetal presentations changed significantly from 20 weeks to delivery, but the presentation at each gestational age interval was significantly associated with the final position of Twin A and the likelihood of vertex-vertex presentation at delivery. The likelihood of spontaneous version of Twin A decreased from 27.9% after 24-27 6/7 weeks, to 18.8% after 28-31 6/7 weeks, to 8.2% after 32-35 6/7 weeks. Pregnancy characteristics associated with spontaneous version of Twin A were a prior vaginal delivery and increased fetal size of either twin. CONCLUSIONS: The likelihood of spontaneous version of Twin A decreases as pregnancy advances. Parity and increased fetal size are associated with spontaneous version.

Objective To determine if prophylactic cerclage improves pregnancy outcomes in women with twin pregnancies without a history of cervical insufficiency.Study Design Women with twin pregnancies who received outpatient preterm labor surveillance services between January 1990 and May 2004 for >/=1 day beginning at < 28.0 weeks' gestation were identified from a database. Patients with previous preterm delivery or a diagnosis of cervical incompetence in a previous or in the index pregnancy were excluded. Twin pregnancies managed with prophylactic cerclage were compared with twin pregnancies in which cerclage was not placed. The primary outcome was incidence of preterm birth prior to 32 weeks. Groups were compared using Fisher exact and Mann-Whitney U test statistics.Results Overall, 8,218 twin pregnancies met inclusion criteria, of which 146 women (1.8%) received prophylactic cerclage. Patients who received prophylactic cerclage had a significantly higher incidence of preterm birth before 32 weeks and infants with lower mean birth weight and longer nursery stays. No significant difference was seen in mean gestational age at delivery. This study had 80% power to detect a 7% reduction in the primary outcome.Conclusion Prophylactic cerclage was not associated with a lower risk of preterm birth and adverse neonatal outcomes in twin pregnancies without history of cervical insufficiency.

OBJECTIVE: To estimate the effectiveness of antepartum surveillance and delivery at 41 weeks in reducing the risk of stillbirth in advanced maternal age (AMA) patients. STUDY DESIGN: Retrospective cohort study of all patients managed in one maternal-fetal medicine practice from June 2005 to May 2012. We included all singleton pregnancies delivered at >/=20 weeks of gestation. All AMA patients (age >/=35 years at their estimated delivery date) underwent weekly biophysical profile testing beginning at 36 weeks, as well as planned delivery at 41 weeks, or sooner if indicated. We compared the rate of fetal death at >/=20 weeks and fetal death at >/=36 weeks in AMA vs. non-AMA patients. Fetal deaths due to lethal and chromosomal abnormalities were excluded. RESULTS: 4469 patients met the inclusion criteria: 1541 (34.5%) were AMA and 2928 (65.5%) were non-AMA. Using our AMA protocol for surveillance and timing of delivery, the incidence of stillbirth was similar to the non-AMA population (stillbirth >/=20 weeks: 3.9 per 1000 vs. 3.4 per 1000, p=0.799; stillbirth >/=36 weeks: 1.4 per 1000 vs. 1.1 per 1000, p=0.773). When looking at women age <35, age 35-39, and age >/=40, the incidence of stillbirth >/=20 weeks and >/=36 weeks did not increase across the three groups. Our findings were similar when we excluded all patients with other indications for antepartum surveillance. CONCLUSIONS: In AMA patients, antepartum surveillance and delivery at 41 weeks appears to reduce the risk of stillbirth to that of the non-AMA population. Routine antepartum surveillance should be considered in all AMA patients.