[Clinical study on the correlation between the duration of labor and postpartum hemorrhage]

Objective: To study the correlation between the duration of labor and postpartum hemorrhage. Methods: The delivery data of singleton first-term pregnant women who delivered vaginally at Beijing Obstetrics and Gynecology Hospital from January 1, 2017 to December 31, 2017 were collected, 3 104 cases met inclusion criteria. According to the duration of the first labor (t), they were divided into two groups: t≥8 hours was the observation group, t<8 hours was the control group. In order to ensure the baseline characteristics of the subjects in the two groups were balanced, propensity score matching (PSM) was adopted, with matching ratio 1∶1. Then the observation group was divided into four subgroups: group 8-12 h, group 12-16 h, group 16-20 h, group ≥20 h. According to the presence or absence of labor intervention (oxytocin use, artificial rupture of membranes, labor analgesia), the observation group and control group were divided into non-labor intervention observation group, non-labor intervention control group, labor intervention observation group, and labor intervention control group. The case data of 3 104 subjects were analyzed and the duration of labor and atonic postpartum hemorrhage rate of each subgroup were compared with the control group. Results: The duration of the second stage of labor and the first+second stages of labor in the observation group (median:0.8, 13.3 hours) and its subgroups were both longer than those in the control group (median:0.6, 5.1 hours), with statistically significant differences (all P<0.01). The rate of atonic postpartum hemorrhage in the observation group, group 16-20 h and group ≥20 h were higher than that in the control group [8.0%(124/1 552), 14.3%(41/287), 14.1%(12/85), 4.6%(72/1 552)], with significant statistical differences (all P<0.01). The duration of the second stage of labor and the first+second stages of labor in the observation group were both longer than those in the control group, regardless of the presence or absence of labor intervention, with statistically significant differences (all P<0.01). In both the observation group and the control group, the duration of the first stage of labor, the second stage of labor, and the first+second stages of labor with labor intervention were longer than those of the non-labor intervention, with significant statistical differences (all P<0.01). The rate of atonic postpartum hemorrhage in the observation group with labor intervention [8.7%(110/1 263)] was higher than that in the observation group without labor intervention [4.8%(14/289)], with a statistical difference (P<0.05). Conclusions: With the increase of the duration of the first stage of labor, the rate of atonic postpartum hemorrhage increases. The first stage of labor is closely related to the second stage of labor, and to a certain extent the duration of the second stage of labor increases with the length of the first stage of labor. With the increase of the duration of the first stage of labor, the rate of labor intervention and atonic postpartum hemorrhage also increase, which could serve as a clinical warning that excessive labor intervention may indicate a higher incidence of atonic postpartum hemorrhage.