Closed- or open-glottis pushing for vaginal delivery: a planned secondary analysis of the TRAnexamic Acid for Preventing postpartum hemorrhage after vaginal delivery study

Abstract

Background: The effect on obstetrical outcomes of closed- or open-glottis pushing is uncertain among both nulliparous and parous women.

Objective: This study aimed to assess the association between open- or closed-glottis pushing and mode of delivery after an attempted singleton vaginal birth at or near term.

Study design: This was an ancillary planned cohort study of the TRAAP (TRAnexamic Acid for Preventing postpartum hemorrhage after vaginal delivery) randomized controlled trial, conducted in 15 French maternity units from 2015 to 2016 that enrolled women with an attempted singleton vaginal delivery after 35 weeks' gestation. After randomization, characteristics of labor and delivery were prospectively collected, with special attention to active second-stage pushing and a specific planned questionnaire completed immediately after birth by the attending care provider. The exposure was the mode of pushing, classified into 2 groups: closed- or open-glottis. The main endpoint was operative vaginal delivery. Secondary endpoints were items of maternal morbidity, including severe perineal laceration, episiotomy, postpartum hemorrhage, duration of the second stage of labor, and a composite severe neonatal morbidity outcome. We also assessed immediate maternal satisfaction, experience of delivery, and psychological status 2 months after delivery. The associations between mode of pushing and outcome were analyzed by multivariate logistic regression to control for confounding bias, with multilevel mixed-effects analysis, and a random intercept for center.

Results: Among 3041 women included in our main analysis, 2463 (81.0%) used closed-glottis pushing and 578 (19.0%) open-glottis pushing; their respective operative vaginal delivery rates were 19.1% (n=471; 95% confidence interval, 17.6-20.7) and 12.5% (n=72; 95% confidence interval, 9.9-15.4; P<.001). In an analysis stratified according to parity and after controlling for available confounders, the rate of operative vaginal delivery did not differ between the groups among nulliparous women: 28.7% (n=399) for the closed-glottis and 27.5% (n=64) for the open-glottis group (adjusted odds ratio, 0.93; 95% confidence interval, 0.65-1.33; P=.7). The operative vaginal delivery rate was significantly lower for women using open- compared with closed-glottis pushing in the parous population: 2.3% (n=8) for the open- and 6.7% (n=72) for the closed-glottis groups (adjusted odds ratio, 0.43; 95% confidence interval, 0.19-0.90; P=.03). Other maternal and neonatal outcomes did not differ between the 2 modes of pushing among either the nulliparous or parous groups.

Conclusion: Among nulliparous women with singleton pregnancies at term, the risk of operative vaginal birth did not differ according to mode of pushing. These results will inform shared decision-making about the mode of pushing during the second stage of labor.

Keywords: Valsalva pushing; closed-glottis pushing; maternal satisfaction; open-glottis pushing; operative vaginal delivery; perineal tears; physiological labor; postpartum hemorrhage; second stage of labor; vaginal birth.