Faculty Bibliography

OBJECTIVE:To identify maternal and peripartum characteristics in twin gestations that are associated with postpartum hemorrhage (PPH) in which one or more units of packed red blood cells (PRBCs) were either administered or recommended but declined (PPH + PRBC). METHODS:This retrospective cohort study evaluated all women with twin gestations who delivered at greater than 23 weeks of gestational age at a single, tertiary medical center from 2011 to 2016. Patients were included if they had documentation of estimated blood loss (EBL) at delivery and complete inpatient medical records available for review. Patients with incomplete records or an intrauterine fetal demise of one or both twins were excluded. The primary outcome was PPH + PRBC. Secondary outcomes included PPH with delivery EBL ≥1500 ml, PPH with atony and uterotonic administration, PPH with maternal hemorrhagic morbidity (MHM), and PPH with severe maternal morbidity (SMM). MHM was a composite outcome defined as PPH associated with any of the following: atony requiring uterotonics, any PRBC transfusion (≥1 unit), uterine or hypogastric artery ligation, hysterectomy, compression sutures, intrauterine balloon tamponade, uterine artery embolization, and/or exploratory laparotomy. SMM was a composite outcome defined as PPH associated with any of the following: administration of ≥4 units of PRBC, administration of ≥2 units of PRBC, and ≥2 units of fresh frozen plasma (FFP), return to operating room for any major procedure (excludes dilation and curettage), any peripartum hysterectomy, uterine artery embolization, intrauterine balloon tamponade or compression suture placed and administration of ≥2 units of PRBC, and/or intensive care unit (ICU) admission for invasive monitoring/treatment. A multivariable logistic regression analysis was performed. RESULTS:A total of 1081 women with twin gestations were included. PPH + PRBC occurred in 4.4% (n = 48), delivery EBL ≥1500 ml occurred in 3.9% (n = 42), and atony with uterotonic administration occurred in 12.1% (n = 131) of the study population. The rate of MHM and SMM were 13.9% (n = 150) and 1.9% (n = 20), respectively. Although the rate of cesarean delivery was high overall (83.2%), it was nearly universal in the PPH + PRBC group (97.9%; p < .02). PPH + PRBC occurred at a rate of 0.5% (n = 1/182) among vaginally delivered twins compared to 5.2% (n = 47/899) among those delivered by cesarean (p < .03). The final multivariable logistic regression model to predict PPH + PRBC identified six significant maternal and peripartum factors: nulliparity, either pregestational or gestational diabetes, intrapartum magnesium sulfate, admission hematocrit <30%, admission platelets <100 000/µL and administration of general anesthesia. CONCLUSIONS:A number of maternal and peripartum factors are associated with PPH in twin gestations. Optimization of maternal hematologic parameters and chronic medical conditions, and reduction in the rate of cesarean delivery in twin pregnancies may decrease the risk of postpartum hemorrhage.