Clinical implications of crown-rump length discordance at 11 to 14 weeks in dichorionic twins.

BACKGROUND: Crown-rump length discordance, defined as ≥10% discordance, has been investigated as an early sonographic marker of subsequent growth abnormalities and is associated with an increased risk of fetal loss in twin pregnancies. Previous studies have not investigated the prevalence of fetal aneuploidy or structural anomalies in twins with discordance or the independent association of crown-rump length discordance with adverse perinatal outcomes. Moreover, data are limited on cell-free DNA screening for aneuploidy in dichorionic twins with discordance. OBJECTIVE: This study aimed to evaluate whether crown-rump length discordance in dichorionic twins between 11 and 14 weeks of gestation is associated with a higher risk of aneuploidy, structural anomalies, or adverse perinatal outcomes and to assess the performance of cell-free DNA screening in dichorionic twin pregnancies with crown-rump length discordance. STUDY DESIGN: This was a secondary analysis of a multicenter retrospective cohort study that evaluated the performance of cell-free DNA screening for the common trisomies in twin pregnancies from December 2011 to February 2020. For this secondary analysis, we included live dichorionic pregnancies with crown-rump length measurements between 11 and 14 weeks of gestation. First, we compared twin pregnancies with discordant crown-rump lengths with twin pregnancies with concordant crown-rump lengths and analyzed the prevalence of aneuploidy and fetal structural anomalies in either twin. Second, we compared the prevalence of a composite adverse perinatal outcome, which included preterm birth at <34 weeks of gestation, hypertensive disorders of pregnancy, stillbirth or miscarriage, small-for-gestational-age birthweight, and birthweight discordance. Moreover, we assessed the performance of cell-free DNA screening in pregnancies with and without crown-rump length discordance. Outcomes were compared with multivariable regression to adjust for confounders. RESULTS: Of 987 dichorionic twins, 142 (14%) had crown-rump length discordance. The prevalence of aneuploidy was higher in twins with crown-rump length discordance than in twins with concordance (9.9% vs 3.9%, respectively; adjusted relative risk, 2.7; 95% confidence interval, 1.4-4.9). Similarly, structural anomalies (adjusted relative risk, 2.5; 95% confidence interval, 1.4-4.4]) and composite adverse perinatal outcomes (adjusted relative risk, 1.2; 95% confidence interval, 1.04-1.3) were significantly higher in twins with discordance. A stratified analysis demonstrated that even without other ultrasound markers, there were increased risks of aneuploidy (adjusted relative risk, 3.5; 95% confidence interval, 1.5-8.4) and structural anomalies (adjusted relative risk, 2.7; 95% confidence interval, 1.5-4.8) in twins with CRL discordance. Cell-free DNA screening had high negative predictive values for trisomy 21, trisomy 18, and trisomy 13, regardless of crown-rump length discordance, with 1 false-negative for trisomy 21 in a twin pregnancy with discordance. CONCLUSION: Crown-rump length discordance in dichorionic twins is associated with an increased risk of aneuploidy, structural anomalies, and adverse perinatal outcomes, even without other sonographic abnormalities. Cell-free DNA screening demonstrated high sensitivity and negative predictive values irrespective of crown-rump length discordance; however, 1 false-negative result illustrated that there is a role for diagnostic testing. These data may prove useful in identifying twin pregnancies that may benefit from increased screening and surveillance and are not ascertained by other early sonographic markers.

Cell-free DNA screening for trisomy 21 in twin pregnancy: a large multicenter cohort study.

BACKGROUND: Analysis of cell-free DNA from maternal blood provides effective screening for trisomy 21 in singleton pregnancies. Data on cell-free DNA screening in twin gestations are promising although limited. In previous twin studies, cell-free DNA screening was primarily performed in the second trimester and many studies did not report chorionicity. OBJECTIVE: This study aimed to evaluate the screening performance of cell-free DNA for trisomy 21 in twin pregnancies in a large, diverse cohort. A secondary aim was to evaluate screening performance for trisomy 18 and trisomy 13. STUDY DESIGN: This was a retrospective cohort study of twin pregnancies from 17 centers for which cell-free DNA screening was performed from December 2011 to February 2020 by one laboratory using massively parallel sequencing technology. Medical record review was conducted for all newborns and data on the birth outcome, the presence of any congenital abnormalities, phenotypic appearance at birth, and any chromosomal testing that was undertaken in the antenatal or postnatal period were extracted. Cases with a possible fetal chromosomal abnormality with no genetic test results were reviewed by a committee of maternal-fetal medicine geneticists. Cases with a vanishing twin and inadequate follow-up information were excluded. A minimum of 35 confirmed cases of trisomy 21 was required to capture a sensitivity of at least 90% with a prevalence of at least 1.9% with 80% power. Test characteristics were calculated for each outcome. RESULTS: A total of 1764 samples were sent for twin cell-free DNA screening. Of those, 78 cases with a vanishing twin and 239 cases with inadequate follow-up were excluded, leaving a total of 1447 cases for inclusion in the analysis. The median maternal age was 35 years and the median gestational age at cell-free DNA testing was 12.3 weeks. In total, 81% of the twins were dichorionic. The median fetal fraction was 12.4%. Trisomy 21 was detected in 41 of 42 pregnancies, yielding a detection rate of 97.6% (95% confidence interval, 83.8-99.7). There was 1 false negative and no false positive cases. Trisomy 21 was detected in 38 out of 39 dichorionic twin pregnancies, yielding a detection rate of 97.4% (95% confidence interval, 82.6-99.7). Trisomy 18 was detected in 10 of the 10 affected pregnancies. There was 1 false positive case. Trisomy 13 was detected in 4 of the 5 cases, yielding a detection rate of 80% (95% confidence interval, 11.1-99.2). There was one false negative and no false positive cases. The nonreportable rate was low at 3.9 %. CONCLUSION: Cell-free DNA testing is effective in screening for trisomy 21 in twin gestations from the first trimester of pregnancy. Detection of trisomy 21 was high in dichorionic and monochorionic twins, and the nonreportable result rates were low. This study included high numbers of cases of trisomy 18 and 13 when compared with the current literature. Although screening for these conditions in twins seems to be promising, the numbers were too small to make definitive conclusions regarding the screening efficacy for these conditions. It is possible that cell-free DNA testing performance may differ among laboratories and vary with screening methodologies.

Genetic abnormalities seen on CVS in early pregnancy failure.

Objective: To determine the frequency and distribution of chromosome abnormalities in women with early pregnancy failure (EPF) detected by cytogenetic testing on chorionic villus sampling.Method: Retrospective observational cohort study of chromosomal analysis from transvaginal chorionic villus sampling (CVS) or reflex products of conception (POC) karyotype. CVS was offered as a training tool for Maternal Fetal Medicine fellows prior to manual vacuum aspiration for EPF 9-week gestation. POC were analyzed for cytogenetics if no results were obtained on CVS.Results: One hundred thirty samples were collected from December 2011 to April 2015. 33 (27.3%) cases had a normal karyotype and 88 (73.0%) cases had an abnormal karyotype. The most common group of abnormalities were trisomy, (n = 50, 41.3%), triploidy/tetraploidy, (n = 17, 14.0%), monosomy (n = 15, 12.4%), and structural rearrangements (n = 6, 5.0%). Nine (6.9%) samples were maternal decidua only. Abnormal karyotype in EPF was significantly increased in women by age group (p < .01) but not in women with a history of prior miscarriage (p = .5).Conclusion: Our cohort had a high detection rate of aneuploidy. The most common chromosomal abnormalities in EPF were: trisomy, followed by triploidy/tetraploidy, monosomy, and structural rearrangements. Maternal age had the strongest correlation with EPF associated with aneuploidy.

How many procedures does it take? Success of a CVS training program for Maternal Fetal Medicine fellows.

OBJECTIVES: To quantify the learning curve for a training program for Maternal Fetal Medicine (MFM) fellows in obtaining successful transvaginal chorionic villus sampling (CVS) results in women with early pregnancy failure (EPF). METHODS: Retrospective observational cohort study of transvaginal CVS and subsequent manual vacuum aspiration (MVA) performed by MFM fellows. CVS samples were sent for karyotype, and products of conception (POC) were sent if CVS sample did not yield a result. Success was defined as karyotype result on CVS specimen. RESULTS: A total of 130 women with EPF up to 9 weeks of gestation underwent transvaginal CVS and MVA from December 2011 to April 2015. CVS samples were successful in 53 (40.8%) cases, POC were analyzed for karyotype in 68 (52.3%) cases, and maternal decidua was obtained in 9 (6.9%) cases. Nine MFM fellows performed the CVS and MVA procedures. The mean number of procedures per fellow was 14 (5-24). The average success rate of transvaginal CVS sample was 33.3% after the first procedure and 50% at the 14th procedure. One procedure was performed per patient. CONCLUSIONS: Success increased over time from 33.3% to 50.0%. Given the gestational age and failed pregnancy status, this is a reasonable success rate for CVS at time of EPF. © 2016 John Wiley & Sons, Ltd.

Second Trimester Cervical Length and Its Association with Vaginal Birth after Cesarean Delivery.

OBJECTIVE: This study aims to evaluate the relationship between cervical length (CL) and trial of labor after cesarean (TOLAC) delivery. STUDY DESIGN: This retrospective cohort study evaluated second trimester CL and likelihood of successful TOLAC. Women undergoing TOLAC from January 2009 to December 2011 who underwent transvaginal ultrasound CL between 18(0/7) to 23(6/7) weeks gestation were included. The threshold for CL was assessed using a receiver operating characteristics curve, with ≥ 45 mm determined to be the optimal threshold. Women with CL < 45 mm were compared with women with CL ≥ 45 mm with respect to TOLAC. Multivariate analysis was used to assess the association of CL with TOLAC. RESULTS: A total of 45 patients met inclusion criteria, 14 (31%) with a long CL (≥ 45 mm) and 31 (69%) with CL < 45 mm. Groups did not differ in demographics or factors predictive of vaginal birth after cesarean (VBAC) delivery. More women with CL < 45 mm had successful VBAC (81%) than women with long CL (43%, relative risk 1.9, p = 0.03). Prior vaginal delivery and CL < 45 mm were both significant predictors of VBAC (odds ratio 1.8 and 2.0, respectively). CONCLUSION: CL < 45 mm is associated with VBAC.

Is universal cervical length screening indicated in women with prior term birth?

OBJECTIVE: To determine whether universal transvaginal ultrasound (TVU) cervical length (CL) screening is indicated in women with prior term births. STUDY DESIGN: We conducted a prospective cohort study from Jan. 1, 2012, to June 30, 2013, of singleton gestations between 18 0/7-23 6/7 weeks undergoing TVU CL screening for prediction of spontaneous preterm birth (sPTB). Women with a prior sPTB, with cerclage, and without delivery data available were excluded. Primary outcomes were the incidence of a CL ≤20 mm, and rates of sPTB <37, <34, and <32 weeks gestation among women with prior term births vs nulliparous women. RESULTS: A total of 1569 women underwent TVU CL screening; 18 women with a prior iatrogenic preterm birth were excluded. Of those screened, 756 (48.7%) had a prior term delivery and 795 (51.3%) were nulliparous. Women with prior term births differed from nulliparous women with respect to age, body mass index, race, smoking status, and prior dilation and curettage. After adjustment for these confounders, there was no statistical difference in the incidence of CL ≤20 mm (0.8 vs 1.4%; adjusted odds ratio, 0.35; 95% confidence interval, 0.12-1.03) among women with prior term births compared with nulliparous women. Rates of sPTB <37, <34, <32 weeks were similar in women with a prior term birth compared with nulliparous women. The overall incidence of sPTB <37 weeks was 3.2% in women with a prior term birth and 5.0% in nulliparous women. When a CL ≤20 mm was identified, 3 of 6 women with prior term birth delivered at <37 weeks compared with 8 of 11 nulliparous women. There were no identified differences between the odds of sPTB between women with a CL ≤20 mm who had a prior term birth compared with nulliparous women. CONCLUSION: Women with prior term birth have a trend toward a lower incidence of CL ≤20 mm, compared with nulliparous women. Further studies are needed to determine whether women with prior term births should be included or excluded from TVU CL screening.

Implementation of a universal cervical length screening program for the prevention of preterm birth.

OBJECTIVE: The objective of this article is to evaluate the implementation and acceptability of a universal cervical length (CL) screening program for prediction and prevention of preterm birth (PTB). STUDY DESIGN: We performed a prospective observational study to evaluate the implementation and acceptability of a universal CL screening program. Between January 1, 2012, and December 31, 2012, women with singleton gestations, without a cerclage or prior spontaneous PTB, were offered transvaginal ultrasound (TVU) for CL between 18(0/7) and 23(6/7) weeks' gestation. Sonographers and medical staff received education before implementation. Intervention for a short CL was interpreted according to a standard protocol. On June 1, 2012, our program was modified from "opt-in" to "opt-out." SPSS 20.0 (released 2011, IBM statistics for Windows version 20, IBM Corp., Armonk, NY) was used for analysis. RESULTS: Over 12 months, 1,484 (87%) of 1,706 eligible women were offered CL screening, and 1,119 (75%) were actually screened. Women were more likely to accept CL screening if they were nulliparous versus multiparous (83 vs. 68%, p < 0.001) and if the sonographer was female versus male (83 vs. 42%, p < 0.001). Implementation of an "opt-out" protocol did not increase the overall number of women accepting CL screening compared with an "opt-in" approach (76 vs. 75%, p = 0.81) CONCLUSION: Universal CL screening can be feasibly implemented and is acceptable to most women.

Early pregnancy failure as a training tool for chorionic villus sampling.

OBJECTIVES: The goal of this study is to evaluate the success of a training program in chorionic villus sampling (CVS) of early pregnancy failure (EPF) for maternal-fetal medicine (MFM) fellows. METHODS: We conducted a retrospective review of a new training program in CVS for MFM fellows. Women with EPF up to 13 weeks estimated gestational age were offered inclusion in our program and counseled on alternatives. Transcervical CVS was performed for both fellow education and cytogenetic diagnosis. The primary outcome was a successful diagnosis, and the secondary outcome was cell growth by fellow experience. RESULTS: Thirty-nine patients diagnosed with EPF from December 2011 to March 2013 underwent CVS. Villi obtained via CVS yielded a diagnosis in 62% of cases. CVS samples with successful karyotype had more villi but were otherwise similar. CVS by experience showed a trend towards increased success and villi volume after the first five procedures. Abnormal results were obtained in 72% of cases, 15% of which led to changes in care. CONCLUSION: CVS of EPF is an option for the training of MFM fellows as it was acceptable to most patients, and the majority of CVS specimens yielded a karyotype diagnosis.

Estimation of fetal weight in fetuses with abdominal wall defects: comparison of 2 recent sonographic formulas to the Hadlock formula.

OBJECTIVE: Estimation of fetal weight is particularly challenging in fetuses with abdominal wall defects (AWDs). We sought to compare the accuracy and screening efficiency for intrauterine growth restriction (IUGR) of 2 recent sonographic formulas to those of the Hadlock formula (Am J Obstet Gynecol 1985; 151:333-337) in fetuses with AWDs. METHODS: This was a retrospective cohort study of fetuses with AWDs. Fetuses with sonographically estimated fetal weights (EFWs) within 14 days before delivery were included. Using the individual biometric measurements, EFWs were calculated using the Honarvar (Int J Gynaecol Obstet 2001; 73:15-20; femur length [FL]), Siemer (Ultrasound Obstet Gynecol 2008; 31:397-400; FL, biparietal diameter [BPD], and occipitofrontal diameter), and Hadlock (BPD, head circumference, abdominal circumference, and FL) formulas. The calculated EFWs were adjusted for interval growth between the dates of sonography and delivery using published sonographic fetal growth velocity standards. Accuracy and screening efficiency for IUGR were compared. RESULTS: Seventy-six fetuses were included: 53 with gastroschisis and 23 with omphalocele. The median gestational age at delivery was 36.6 weeks (range, 25.0 to 39.0 weeks). The Siemer formula had the lowest mean percentage error (-2.5% [95% confidence interval (CI), -6.2% to +1.2%]) without systematic bias (P = .182). The Hadlock formula had the highest precision (random error, 11.4%), sensitivity (91%), and accuracy for predicting IUGR (85% [95% CI, 77% to 94%]). CONCLUSIONS: None of the 3 sonographic formulas is ideal for estimating fetal weight in fetuses with AWDs. The Siemer formula should be used when accuracy in the absolute EFW is the goal. For the purpose of making the more clinically relevant diagnosis of IUGR, use of the Hadlock formula is justified.

Predicting adverse neonatal outcomes in fetuses with abdominal wall defects using prenatal risk factors.

OBJECTIVE: The aim of this study was to determine whether prenatal variables can predict adverse neonatal outcomes in fetuses with abdominal wall defects. STUDY DESIGN: A retrospective cohort study that used ultrasound and neonatal records for all cases of gastroschisis and omphalocele seen over a 16-year period. Cases with adverse neonatal outcomes were compared with noncases for multiple candidate predictive factors. Univariable and multivariable statistical methods were used to develop the prediction models, and effectiveness was evaluated using the area under the receiver operating characteristic curve. RESULTS: Of 80 fetuses with gastroschisis, 29 (36%) had the composite adverse outcome, compared with 15 of 33 (47%) live neonates with omphalocele. Intrauterine growth restriction was the only significant variable in gastroschisis, whereas exteriorized liver was the only predictor in omphalocele. The areas under the curve for the prediction models with gastroschisis and omphalocele are 0.67 and 0.74, respectively. CONCLUSION: Intrauterine growth restriction and exteriorization of the liver are significant predictors of adverse neonatal outcome with gastroschisis and omphalocele.

Discussion: 'Hypnotic analgesia during first-trimester termination' by Marc et al.

In the roundtable that follows, clinicians discuss a study published in this issue of the Journal in light of its methodology, relevance to practice, and implications for future research. Article discussed: Marc I, Rainville P, Masse B, et al. Hypnotic analgesia intervention during first-trimester pregnancy termination: an open randomized trial. Am J Obstet Gynecol 2008;199:469.e1-469.e9.

Hypnotic analgesia during first-trimester termination: Marc et al.

The article below summarizes a roundtable discussion of a study published in this issue of the Journal in light of its methodology, relevance to practice, and implications for future research. Article discussed: Marc I, Rainville P, Masse B, et al. Hypnotic analgesia intervention during first-trimester pregnancy termination: an open randomized trial. Am J Obstet Gynecol 2008;199:469.e1-469.e9. The full discussion appears at www.AJOG.org, page e1-e5.