Duration of Induction of Labor for Second-Trimester Medication Abortion and Adverse Outcomes

Abstract

Objective: To evaluate the relationship between duration of labor during second-trimester medication abortion and adverse outcomes.

Methods: We conducted a retrospective cohort study including all individuals with a singleton gestation undergoing second-trimester medication abortion without evidence of advanced cervical dilation, rupture of membranes, or preterm labor at four centers. The primary exposure was duration of labor (ie, hours spent from receiving misoprostol to fetal expulsion). The primary outcome was composite morbidity , defined as uterine rupture, need for blood transfusion, clinical chorioamnionitis, intensive care unit admission, or need for readmission. We performed bivariate and multivariate negative binomial analyses. A post hoc subgroup analysis was performed to assess for the risk of the primary outcome by gestational age. We performed tests of homogeneity based on history of uterine scarring and parity.

Results: Six hundred eighty-one individuals were included. The median duration of labor was 11 hours (interquartile range 8-17 hours). One hundred thirty-one (19.2%) experienced the primary outcome. When duration of labor was evaluated continuously, a longer duration of labor was associated with an increased frequency of morbidity (adjusted β=0.68, 95% CI, 0.32-1.04). When duration of labor was evaluated categorically, those experiencing the highest quartile of duration (ie, 17 hours or more) had a statistically higher risk for experiencing morbidity compared with individuals in all other quartiles (adjusted relative risk 1.99, 95% CI, 1.34-2.96). When we focused on components of the composite outcome, clinical chorioamnionitis was significantly different between those experiencing a longer duration and those experiencing a shorter duration of labor (26.2% vs 10.6%, P <.001). On subgroup analysis, gestational age was not associated with the risk of composite morbidity. Tests of homogeneity demonstrated no significant difference in the risk for morbidity among individuals with a history of uterine scarring or based on parity.

Conclusion: Duration of labor was independently associated with risks for adverse maternal outcomes during second-trimester medication abortion, specifically clinical chorioamnionitis.