Faculty Bibliography

BACKGROUND:The relationship between fetal fraction and birth weight in twin gestations is poorly understood. OBJECTIVE:To investigate the relationship between first trimester cfDNA fetal fraction and birth weight < 10th percentile in twin gestations. STUDY DESIGN/METHODS:This is a planned secondary analysis of the Twin cfDNA Study, a 17-center retrospective cohort of twin pregnancies screened for aneuploidy using cfDNA in the first trimester from 12/2011 - 2/2022, excluding those with positive screen results for chromosomal aneuploidy. CfDNA testing was performed by a single lab using massively parallel sequencing (MPSS). Baseline characteristics and birth weight of pregnancies with normal fetal fraction were compared to those with low (<5%) and high (>95%) fetal fraction using univariable analyses and multivariable regression. RESULTS:A total of 1041 twin pregnancies were included. Chronic hypertension, elevated BMI, and self-identified Black race were associated with fetal fraction <5th percentile. There was no difference in median fetal fraction between those with birth weight <10th percentile in at least one twin (median [IQR] fetal fraction 12.2% [9.8, 14.8] versus those with normal birth weight (10th percentile) in both twins (median [IQR] fetal fraction 12.3% [9.7, 15.2] for normal birth weight, p = 0.49). There was no association between high or low fetal fraction and birth weight <10th percentile for one (p=0.45) or both (p=0.81) twins, and there was no association between high or low fetal fraction and birth weight <5th percentile for one (p=0.44) or both (p=0.74) twins. The results were unchanged after adjustment for potential confounders. CONCLUSION/CONCLUSIONS:In this large cohort, there was no association between the extremes of cfDNA fetal fraction and birthweight < 10th percentile, suggesting that first trimester fetal fraction may not predict impaired fetal growth in twin gestations.

OBJECTIVE: We sought to create a machine learning (ML) model to identify variables that would aid in the prediction of surgical morbidity in cases of placenta accreta spectrum (PAS). STUDY DESIGN/METHODS: A multicenter analysis including all cases of PAS identified by pathology specimen confirmation, across five tertiary care perinatal centers in New York City from 2013 to 2022. We developed models to predict operative morbidity using 213 variables including demographics, obstetrical information, and limited prenatal imaging findings detailing placental location. Our primary outcome was prediction of a surgical morbidity composite defined as including any of the following: blood loss (>1,500 mL), transfusion, intensive care unit admission, vasopressor use, mechanical ventilation/intubation, and organ injury. A nested, stratified, cross-validation approach was used to tune model hyperparameters and estimate generalizability. Gradient boosted tree classifier models incorporated preprocessing steps of standard scaling for numerical variables and one-hot encoding for categorical variables. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), positive and negative predictive values (PPV, NPV), and F1 score. Variable importance ranking was also determined. RESULTS: Among 401 PAS cases, 326 (81%) underwent hysterectomy. Of the 401 cases of PAS, 309 (77%) had at least one event defined as surgical morbidity. Our predictive model had an AUC of 0.79 (95% confidence interval: 0.69, 0.89), PPV 0.79, NPV 0.76, and F1 score of 0.88. The variables most predictive of surgical morbidity were completion of a hysterectomy, prepregnancy body mass index (BMI), absence of a second trimester ultrasound, socioeconomic status zip code, BMI at delivery, number of prenatal visits, and delivery time of day. CONCLUSION/CONCLUSIONS: By identifying social and obstetrical characteristics that increase patients' risk, ML models are useful in predicting PAS-related surgical morbidity. Utilizing ML could serve as a foundation for risk and complexity stratification in cases of PAS to optimize surgical planning. KEY POINTS/CONCLUSIONS:· ML models are useful models are useful in predicting PAS-related surgical morbidity.. · Optimal management for PAS remains unclear.. · Utilizing ML can serve as a foundation for risk and complexity stratification in cases of PAS..

OBJECTIVE: We sought to create a machine learning (ML) model to identify variables that would aid in the prediction of surgical morbidity in cases of placenta accreta spectrum (PAS). STUDY DESIGN/METHODS: A multicenter analysis including all cases of PAS identified by pathology specimen confirmation, across five tertiary care perinatal centers in New York City from 2013 to 2022. We developed models to predict operative morbidity using 213 variables including demographics, obstetrical information, and limited prenatal imaging findings detailing placental location. Our primary outcome was prediction of a surgical morbidity composite defined as including any of the following: blood loss (>1,500 mL), transfusion, intensive care unit admission, vasopressor use, mechanical ventilation/intubation, and organ injury. A nested, stratified, cross-validation approach was used to tune model hyperparameters and estimate generalizability. Gradient boosted tree classifier models incorporated preprocessing steps of standard scaling for numerical variables and one-hot encoding for categorical variables. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), positive and negative predictive values (PPV, NPV), and F1 score. Variable importance ranking was also determined. RESULTS: Among 401 PAS cases, 326 (81%) underwent hysterectomy. Of the 401 cases of PAS, 309 (77%) had at least one event defined as surgical morbidity. Our predictive model had an AUC of 0.79 (95% confidence interval: 0.69, 0.89), PPV 0.79, NPV 0.76, and F1 score of 0.88. The variables most predictive of surgical morbidity were completion of a hysterectomy, prepregnancy body mass index (BMI), absence of a second trimester ultrasound, socioeconomic status zip code, BMI at delivery, number of prenatal visits, and delivery time of day. CONCLUSION/CONCLUSIONS: By identifying social and obstetrical characteristics that increase patients' risk, ML models are useful in predicting PAS-related surgical morbidity. Utilizing ML could serve as a foundation for risk and complexity stratification in cases of PAS to optimize surgical planning. KEY POINTS/CONCLUSIONS:· ML models are useful models are useful in predicting PAS-related surgical morbidity.. · Optimal management for PAS remains unclear.. · Utilizing ML can serve as a foundation for risk and complexity stratification in cases of PAS..

OBJECTIVE: We sought to create a machine learning (ML) model to identify variables that would aid in the prediction of surgical morbidity in cases of placenta accreta spectrum (PAS). STUDY DESIGN/METHODS: A multicenter analysis including all cases of PAS identified by pathology specimen confirmation, across five tertiary care perinatal centers in New York City from 2013 to 2022. We developed models to predict operative morbidity using 213 variables including demographics, obstetrical information, and limited prenatal imaging findings detailing placental location. Our primary outcome was prediction of a surgical morbidity composite defined as including any of the following: blood loss (>1,500 mL), transfusion, intensive care unit admission, vasopressor use, mechanical ventilation/intubation, and organ injury. A nested, stratified, cross-validation approach was used to tune model hyperparameters and estimate generalizability. Gradient boosted tree classifier models incorporated preprocessing steps of standard scaling for numerical variables and one-hot encoding for categorical variables. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), positive and negative predictive values (PPV, NPV), and F1 score. Variable importance ranking was also determined. RESULTS: Among 401 PAS cases, 326 (81%) underwent hysterectomy. Of the 401 cases of PAS, 309 (77%) had at least one event defined as surgical morbidity. Our predictive model had an AUC of 0.79 (95% confidence interval: 0.69, 0.89), PPV 0.79, NPV 0.76, and F1 score of 0.88. The variables most predictive of surgical morbidity were completion of a hysterectomy, prepregnancy body mass index (BMI), absence of a second trimester ultrasound, socioeconomic status zip code, BMI at delivery, number of prenatal visits, and delivery time of day. CONCLUSION/CONCLUSIONS: By identifying social and obstetrical characteristics that increase patients' risk, ML models are useful in predicting PAS-related surgical morbidity. Utilizing ML could serve as a foundation for risk and complexity stratification in cases of PAS to optimize surgical planning. KEY POINTS/CONCLUSIONS:· ML models are useful models are useful in predicting PAS-related surgical morbidity.. · Optimal management for PAS remains unclear.. · Utilizing ML can serve as a foundation for risk and complexity stratification in cases of PAS..

We present a case of a vein of Galen aneurysmal malformation (VGAM), a rare congenital arteriovenous malformation, in one fetus of a monochorionic-diamniotic twin pregnancy. The diagnosis was made with color Doppler ultrasonography at 28 weeks and the affected fetus was found to have worsening cardiomegaly on subsequent fetal echocardiograms. She was emergently delivered at 32 weeks for abnormal fetal heart rate tracing of the affected twin. Magnetic resonance imaging of the brain findings after delivery demonstrated severe neurological injury; therefore, postnatal embolization was not performed. The neonate died on day of life 9. The cotwin survived without neurological complications. This is the first case in the literature of a VGAM diagnosed prenatally in a monochorionic-diamniotic twin pregnancy and demonstrates the challenge of delivery timing with prenatal diagnosis in a twin pregnancy.

BACKGROUND:Placenta accreta spectrum is associated with significant maternal and neonatal morbidity and mortality. There is limited established data on healthcare inequities in the outcomes of patients with placenta accreta spectrum. OBJECTIVE:This study aimed to investigate health inequities in maternal and neonatal outcomes of pregnancies with placenta accreta spectrum. STUDY DESIGN:This multicentered retrospective cohort study included patients with a histopathological diagnosis of placenta accreta spectrum at 4 regional perinatal centers between January 1, 2013, and June 30, 2022. Maternal race and ethnicity were categorized as either Hispanic, non-Hispanic Black, non-Hispanic White, or Asian or Pacific Islander. The primary outcome was a composite adverse maternal outcome: transfusion of ≥4 units of packed red blood cells, vasopressor use, mechanical ventilation, bowel or bladder injury, or mortality. The secondary outcomes were a composite adverse neonatal outcome (Apgar score of <7 at 1 minute, morbidity, or mortality), gestational age at placenta accreta spectrum diagnosis, and planned delivery by a multidisciplinary team. Multivariable logistic regression was used to estimate the associations of race and ethnicity with maternal and neonatal outcomes. RESULTS:A total of 408 pregnancies with placenta accreta spectrum were included. In 218 patients (53.0%), the diagnosis of placenta accreta spectrum was made antenatally. Patients predominantly self-identified as non-Hispanic White (31.6%) or non-Hispanic Black (24.5%). After adjusting for institution, age, body mass index, income, and parity, there was no difference in composite adverse maternal outcomes among the racial and ethnic groups. Similarly, adverse neonatal outcomes, gestational age at prenatal diagnosis, rate of planned delivery by a multidisciplinary team, and cesarean hysterectomy were similar among groups. CONCLUSION:In our multicentered placenta accreta spectrum cohort, race and ethnicity were not associated with inequities in composite maternal or neonatal morbidity, timing of diagnosis, or planned multidisciplinary care. This study hypothesized that a comparable incidence of individual risk factors for perinatal morbidity and geographic proximity reduces potential inequities that may exist in a larger population.

BACKGROUND:Biologically active cervical glands provide a mucous barrier while influencing the composition and biomechanical strength of the cervical extracellular matrix. Cervical remodeling during ripening may be reflected as loss of the sonographic cervical gland area. As sonographic cervical length remains suboptimal for universal screening, adjunctive evaluation of other facets of the mid-trimester cervix may impart additional screening benefit. OBJECTIVE:To sonographically assess the cervical gland area at universal cervical length screening for preterm birth. STUDY DESIGN/METHODS:We performed a retrospective cohort study of singletons with transvaginal cervical length screening universally performed during anatomic survey between 18 0/7 and 23 6/7 weeks and subsequent live delivery at a single institution in 2018. Uterine anomalies, cerclage, suboptimal imaging, or medically indicated preterm birth were excluded. Ultrasound images were assessed for cervical length and cervical gland area (with quantitative measurements when present). The primary outcome was spontaneous preterm birth <37 weeks. Absent and present gland groups were compared using χ2, Fisher's exact, T-test, and multivariate logistic regression (adjusting for parity and progesterone use, as well as the gestational age, cervical length, and gland absence at screening ultrasound). Gland measurements were evaluated using the Mann-Whitney-U Test and Spearman's correlation. RESULTS:Among the cohort of 772 patients, absent and present CGA groups were overall similar. Patients were on average 33 years old, ∼20 weeks gestation at screening ultrasound, and overall, 2.5% had history of prior spontaneous preterm birth. The absent gland group was more likely to have been taking progesterone (17% vs 4%, p=0.04). Overall rate of preterm birth was 2.6%. However, the 2.3% of patients with absent cervical gland area were significantly more likely to deliver <37 weeks (aOR 23.9, 95% CI 6.4-89, p<0.001). Multivariate logistic regression demonstrated better performance of a cervical length screening model for preterm birth prediction with the addition of qualitative gland evaluation (p<0.001). Qualitative gland assessment was reproducible (PABAK 0.89), but quantitative gland measurements did not correlate with preterm birth. CONCLUSION/CONCLUSIONS:Qualitative gland absence at mid-gestation cervical length screening was associated with subsequent spontaneous preterm birth, whereas quantitative gland measurements were not. Multifaceted ultrasound screening may be needed to adequately evaluate the multiple biologic functions of the cervix.

OBJECTIVE:Evidence is inconsistent regarding grand multiparity and its association with adverse obstetric outcomes. Few large American cohorts of grand multiparas have been studied. We assessed if increasing parity among grand multiparas is associated with increased odds of adverse perinatal outcomes. STUDY DESIGN/METHODS:Multi-center retrospective cohort of patients with parity ≥5 who delivered a singleton gestation in New York City from 2011-2019. Outcomes included postpartum hemorrhage, preterm delivery, hypertensive disorders of pregnancy, shoulder dystocia, birthweight >4000 grams and <2500 grams, and NICU admission. Parity was analyzed continuously, and multivariate analysis determined if increasing parity and other obstetric variables were associated with each adverse outcome. RESULTS:There were 2,496 patients who met inclusion criteria. Increasing parity among grand multiparas was not associated with any of the pre-specified adverse outcomes. Odds of postpartum hemorrhage increased with history (aOR 2.65 [1.83, 3.84]) and current cesarean delivery (aOR 4.59 [3.40, 6.18]). Preterm delivery was associated with history (aOR 12.36 [8.70-17.58]) and non-White race (aOR 1.90 [1.27, 2.84]). Odds of shoulder dystocia increased with history (OR 5.89 [3.22, 10.79]) and birth weight >4000g (aOR 9.94 [6.32, 15.65]). Birthweight >4000 grams was associated with maternal obesity (aOR 2.92 [2.22, 3.84]). Birthweight <2500 grams was associated with advanced maternal age (aOR 1.69 [1.15, 2.48]), chronic hypertension (aOR 2.45 [1.32, 4.53]) and non-White race (aOR 2.47 95% CI [1.66, 3.68]). Odds of hypertensive disorders of pregnancy increased with advanced maternal age (aOR 1.79 [1.25, 2.56]), history (aOR 10.09 [6.77-15.04]) and non-White race (aOR 2.79 [1.95, 4.00]). NICU admission was associated with advanced maternal age (aOR 1.47 [1.06, 2.02]) and non-White race (aOR 2.57 [1.84, 3.58]). CONCLUSION/CONCLUSIONS:Among grand multiparous patients, the risk factor for adverse maternal, obstetric and neonatal outcomes, appears to be occurrence of those adverse events in a prior pregnancy and not increasing parity itself.

KEY WORDS/BACKGROUND:postpartum hemorrhage, thromboelastography, coagulopathy, fibrinogen, massive transfusion. OBJECTIVE:Thromboelastography, a point-of-care test that measures blood's dynamic viscoelastic properties, is routinely used to guide resuscitation in surgical specialties with high hemorrhage risk. Patients with ongoing postpartum hemorrhage often develop coagulopathy and hypofibrinogenemia. Timely assessment of fibrinogen is crucial because cryoprecipitate for repletion requires thawing prior to administration. Thromboelastography may provide rapid assessment of coagulopathy in ongoing hemorrhage but this has not been thoroughly studied. Our objective was to determine if thromboelastography accurately reflects coagulopathy in ongoing postpartum hemorrhage when compared to standard assays. STUDY DESIGN/METHODS:This was a retrospective cohort study of people with ongoing postpartum hemorrhage (quantified blood loss >1000 mL), from 1/1/16-12/31/19. Thromboelastography variables and standard coagulation parameters were compared in patients who had both assays drawn simultaneously. As a secondary analysis, patients who had thromboelastography were compared to those who did not. Mann-Whitney, Fisher's Exact, Kruskal-Wallis, Spearman's Rho, and logistic regression tests were used for analysis. Significance was set at p < 0.05. RESULTS:A total of 680 patients were included. 69 had thromboelastography and coagulation parameters drawn simultaneously and were included in the primary analysis. The remainder were included in the secondary analysis. Thromboelastography variables and coagulation assays correlated significantly - prolonged R with increased PTT (rho 0.25, p=0.04), prolonged K and decreased alpha angle with decreased fibrinogen (rho -0.61, p<0.001; rho 0.24, p<0.001), and decreased maximum amplitude with decreased platelets (rho 0.62, p<0.001). Those who had thromboelastographic assays had higher blood loss and need for interventions to manage hemorrhage than those who did not. CONCLUSION/CONCLUSIONS:Thromboelastography correlated significantly with standard laboratory assays in ongoing postpartum hemorrhage, including for patients with hypofibrinogenemia Given the point-of-care nature and rapid turnaround time, thromboelastography should be considered for timely hemorrhage evaluation and directed resuscitation of coagulopathy.