Objective
The objective of the study was to determine the relationships between maternal race and obstetric outcomes in twin gestations by planned mode of delivery.
Study Design
We performed a secondary analysis of the Consortium on Safe Labor data. Patients with twin gestations in vertex-vertex presentation greater than 32 weeks’ gestational age were grouped according to race. Demographic information and neonatal and maternal outcomes were analyzed according to planned mode of delivery: elective cesarean or trial of labor (with subsequent vaginal delivery, unplanned cesarean, or combined delivery). The primary outcome was unplanned cesarean. Secondary outcomes included maternal and neonatal outcomes.
Results
One thousand nine vertex-vertex twin pregnancies were identified. There were no significant differences across ethnicities in the rate of unplanned cesarean delivery, which occurred in 233 of patients undergoing trial of labor (27%). Elective cesarean occurred in 151 patients (15%). African American women were less likely to have an elective cesarean compared with whites (odds ratio, 0.5; 95% confidence interval, 0.3–0.8), and Asian women were more likely to have an elective cesarean compared with whites (odds ratio, 2.0; 95% confidence interval, 1.2–3.4. Combined delivery occurred in 67 patients (8%) and did not differ among the groups. Subgroup analysis did not reveal any significant differences in neonatal outcomes. Adverse maternal outcomes were rare across ethnicities.
Conclusion
Unplanned cesarean delivery rates are similar in twin pregnancies, regardless of race. Maternal and neonatal outcomes in twin gestations are similar across ethnicities, regardless of mode of delivery.
The rise in prevalence of multiple gestations has been well documented. Although they account for only 3% of all live births, multiple gestations are responsible for a disproportionate amount of maternal and neonatal morbidity and mortality. Despite extensive research, mode of delivery for twin gestations remains a controversial area of management. Most publications and American Congress of Obstetricians and Gynecologists guidelines support a trial of labor for twins in vertex-vertex presentation ; however, 25% of patients opt for cesarean delivery despite vertex-vertex presentation.
A recent metaanalysis reported on 18 studies, most using birth registry data, examining neonatal outcomes for 39,571 twin sets by birth order, presentation, and planned and actual mode of delivery. This study concluded that outcomes were better for twin A than twin B, no differences were noted between vertex and nonvertex presentations for twin B, and no differences were noted between attempted vaginal vs planned cesarean delivery. In twin A, neonatal morbidity was lower after a vaginal delivery. In twin B, morbidity following a combined delivery (vaginal delivery of twin A, followed by cesarean delivery of twin B) was higher than after a vaginal delivery or planned cesarean.
Racial disparities in perinatal outcomes have been well documented. It is known that African American race is a significant risk factor for many adverse pregnancy outcomes, including preterm delivery, preeclampsia, and other obstetric complications such as perinatal death. Racial and ethnic disparities in rates of primary unplanned cesarean delivery in the singleton population also have been reported.
There is paucity of data on racial disparities in neonatal and maternal outcomes in twins. It is unclear whether certain ethnic groups are at higher risk for unplanned cesarean and whether mode of delivery has a role in preventing neonatal morbidity and mortality. Our aim was to determine relationships between maternal race and obstetric outcomes in twin gestations stratified by planned mode of delivery, while controlling for gestational age, birth order, and other confounding variables. We surmised that if racial disparities were found in twin outcomes by mode of delivery, interventions could be targeted to reduce disparities and therefore improve overall perinatal outcomes.
Our hypothesis was that there are no significant differences in the rates of unplanned cesarean delivery and that maternal and neonatal outcomes are similar across ethnicities. Establishing the safety of a trial of labor in vertex-vertex twins across ethnicities may encourage patients and providers to attempt a trial of labor, which could contribute to decreasing the primary cesarean section rate.
Materials and Methods
This is a secondary analysis of prospectively collected data from the Consortium on Safe Labor, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health from 2002 to 2008. The Consortium on Safe Labor contains obstetric and neonatal data from electronic medical records of multiple medical centers throughout the United States, including 19 hospitals at 12 institutions, representing university- and community-based practices. Centers were selected based on numerous criteria including their geographic location (9 states and the District of Columbia).
All deliveries with a gestation of 23 weeks or longer were included in the database. Maternal and neonatal information, as recorded in the electronic medical records, included maternal demographics, reproductive and medical history, prenatal history of current pregnancy, labor and delivery information, and newborn outcomes. The protocol was approved by the institutional review board in all participating institutions for the prospective collection of data for the Consortium on Safe Labor database. This is a secondary analysis of deidentified data and was therefore found to be exempt of institutional review board review by the MedStar Health Research Institute Institutional Review Board.
Patients with twin gestations, greater than 32 weeks estimated gestational age, and in vertex-vertex presentation at delivery were grouped according to race (white, African American, Hispanic, Asian, or other/unknown), which was self-reported. We excluded patients with a history of a previous cesarean delivery, fetal congenital anomalies, and patients with a contraindication to a vaginal delivery such as placenta or vasa previa.
Demographic information and neonatal and maternal outcomes were extracted from the database and analyzed according to planned mode of delivery: elective cesarean delivery or planned vaginal delivery (trial of labor). The elective cesarean delivery group included patients who may have had an indication for delivery prior to their planned cesarean date, such as preterm labor or preeclampsia, but had no contraindication to a vaginal delivery. The planned vaginal delivery group was further stratified into the actual route of delivery: successful vaginal delivery, unplanned cesarean, or combined delivery (defined as vaginal delivery for twin A followed by unplanned cesarean for twin B).
The primary outcome was unplanned cesarean section in the trial of labor group. Secondary outcomes included maternal outcomes: postpartum hemorrhage (defined as estimated blood loss >500 mL for a vaginal delivery and >1000 mL for a cesarean), blood transfusion, intensive care unit (ICU) admission, repeat laparotomy, and maternal death; and neonatal outcomes: 5 minute Apgar score less than 7, neonatal intensive care unit (NICU) admission stratified by gestational age, NICU length of stay longer than 4 days, and intrapartum death. Because prematurity is strongly related to NICU admission, the outcome of NICU admission was evaluated in a subgroup analysis: preterm deliveries (32 to 36 6/7 weeks), and term deliveries (≥37 weeks).
Multivariate logistic regression was performed for the outcomes of elective and unplanned cesarean delivery. The following confounding variables were controlled for: gestational age, parity, clinical sites, age, body mass index, induction of labor, education, and use of assisted reproductive technologies (ART). The variable of ART was not reported by 7 of the 19 participating hospitals, so 2 regressions were performed: 1 using the full dataset without an adjustment for ART and 1 with data restricted to those centers that reported ART and that included ART as a covariate. Because the results were similar, final results are presented using the data from the full data set and without adjustment for ART. Missing data are presented in Table 1 . Statistical analysis was performed using χ 2 , with statistical significance set at P < .05. Analyses were conducted using SAS version 9.1.3 (SAS Institute, Inc, Cary, NC).
| Variable | White (n = 543) a | African American (n = 274) | Hispanic (n = 131) | Asian (n = 31) | Other or unknown (n = 30) | P value |
|---|---|---|---|---|---|---|
| Percentage of the population | 54 | 27 | 13 | 3 | 3 | NA |
| Age (SD) | 32.0 (6.4) | 26.6 (6.6) | 27.6 (6.4) | 32 (6.1) | 28.7 (5.8) | .8 |
| BMI | 31 (6.1) | 34 (7.3) | 32 (7) | 29 (2.8) | 34 (7.9) | < .001 |
| Nulliparity n (%) | 304 (56) | 104 (38) | 56 (43) | 23 (74) | 16 (53) | .42 |
| Education, n (%) b | 364 (67) | 88 (32) | 25 (19) | 25 (8) | 6 (2) | .01 |
| ART n (%) (missing data: 265) | 119 (22) | 14 (5) | 4 (3) | 8 (26) | 1 (3) | .01 |
| IOL, n (%) | 168 (31) | 104 (38) | 45 (34) | 9 (30) | 7 (23) | .64 |
a n is the total number of patients in each subgroup
A post hoc analysis was performed to determine whether the sample size was adequate to find a difference if such a difference exists or whether the lack of a statistically significant difference could be caused by a type 2 or beta error. A sample size of 138 in each group was required to achieve 80% power to detect a 15% difference between the groups, for a baseline outcome rate of 20%. We were therefore adequately powered to find a difference in the primary outcome of unplanned cesarean section and the secondary outcome of NICU admission. The statistical test used the 2-sided Z test with pooled variance. The significance level of the test was targeted at 95%.
Results
There were 228,668 deliveries in the database, 4846 of which were twin gestations, 4607 at an estimated gestational age of 32 weeks or longer. We excluded patients with a prior uterine scar (n = 635), pregnancies complicated by placenta previa or placental abruption (n = 165), presentations other than vertex-vertex (n = 2492), active herpes virus or human immunodeficiency virus infection (n = 10), fetal anomalies (n = 251), and missing data for race (n = 45), for a total sample size of 1009. Demographic characteristics are presented in Table 1 .
There was no statistically significant difference across ethnicities with respect to rate of unplanned cesarean delivery, which was performed for 233 of the 858 women undergoing a trial of labor (27%). Vaginal delivery occurred in 558 patients (65%), and combined delivery (unplanned cesarean delivery of twin B after vaginal delivery of twin A) occurred in 67 (8%) patients ( Table 2 ).
| Variable | n (%) | OR (95% CI) | aOR (95% CI) |
|---|---|---|---|
| Unplanned cesarean delivery | |||
| White | 128 (28) | Referent | Referent |
| African American | 64 (25) | 1.0 (0.8–1.3) | 0.9 (0.5–1.4) |
| Hispanic | 24 (23) | 0.8 (0.5–1.2) | 0.7 (0.4–1.3) |
| Asian | 8 (38) | 1.1 (0.6–2.0) | 1.7 (0.6–4.3) |
| Other | 9 (33) | 1.3 (0.7–2.2) | 1.2 (0.4–2.9) |
| Vaginal delivery | |||
| White | 291 (63) | Referent | Referent |
| African American | 165 (65) | 1.1 (1.0–1.3) | 1.2 (0.8–2.0) |
| Hispanic | 74 (72) | 1.1 (0.9–1.2) | 1.5 (0.8–2.6) |
| Asian | 13 (62) | 0.8 (0.5–1.2) | 0.8 (0.3–2.1) |
| Other | 15 (56) | 0.9 (0.6–1.3) | 0.7 (0.3–1.6) |
| Combined delivery | |||
| White | 36 (8) | Referent | Referent |
| African American | 23 (9) | 1.3 (0.8–2.1) | 0.9 (0.4–1.9) |
| Hispanic | 5 (5) | 0.6 (0.2–1.4) | 0.8 (0.3–2.4) |
| Asian | 0 (0) | NA | NA |
| Other | 3 (11) | 1.5 (0.5–4.6) | 2.0 (0.5–7.8) |
African American women were less likely to have an elective cesarean delivery compared with other ethnic groups (odds ratio, 0.5; 95% confidence interval, 0.32–0.8); P < .001), while Asian women were most likely to undergo an elective cesarean (OR, 2.0; 95% confidence interval, 1.2–3.4; P = .01; Table 3 ).
| Demographic | n (%) | OR (95% CI) | aOR (95% CI) |
|---|---|---|---|
| White | 88 (16) | Referent | Referent |
| African American | 22 (8) | 0.5 (0.3–0.8) | 0.7 (0.3–0.9) |
| Hispanic | 28 (21) | 1.3 (0.9–1.9) | 2.1 (0.9–3.6) |
| Asian | 10 (32) | 2 (1.2–3.4) | 2.7 (1.1–6.2) |
| Other | 3 (10) | 0.6 (0.2–1.8) | 0.9 (0.3–3.4) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
