Importance: Amniotic fluid embolism (AFE) is an uncommon pregnancy complication but is associated with high maternal mortality. Because of the rarity of AFE, associated risks factors and maternal outcomes have been relatively understudied.
Objective: To examine the clinical, pregnancy, and delivery characteristics and the maternal outcomes related to AFE in a recent period in the US.
Design, setting, and participants: This retrospective cohort study examined hospital deliveries from January 1, 2016, to December 31, 2019, from the Healthcare Cost and Utilization Project's National Inpatient Sample.
Main outcomes and measures: The primary outcome was clinical, pregnancy, and delivery characteristics of AFE, assessed with a multivariable binary logistic regression model. The coprimary outcome was failure to rescue, defined as maternal mortality after AFE. Associations with other severe maternal morbidity indicators and failure to rescue per clinical and pregnancy characteristics were also assessed.
Results: A total of 14 684 135 deliveries were examined, with AFE diagnosed in 880 women, corresponding to an incidence rate of 6.0 per 100 000 deliveries. The cohort-level median patient age was 29 years (IQR, 25-33 years). In a multivariable analysis, (1) patient factors of older age, Asian and Black race, Western US region, pregestational hypertension, asthma, illicit substance use, and grand multiparity; (2) pregnancy factors of placental accreta spectrum (PAS), placental abruption, uterine rupture, polyhydramnios, chorioamnionitis, preeclampsia, fetal growth restriction, and fetal demise; and (3) delivery factors of early gestational age, cervical ripening, cesarean delivery, operative delivery, and manual removal were associated with AFE. Among these characteristics, PAS had the largest association with AFE (adjusted odds ratio [aOR], 10.01; 95% CI, 7.03-14.24). When stratified by the PAS subtypes, more severe forms of PAS had a greater association with AFE (aOR for increta and percreta, 17.35; 95% CI, 10.21-28.48; and aOR for accreta, 7.62; 95% CI, 4.83-12.01). Patients who had AFE were more likely to have coagulopathy (aOR, 24.68; 95% CI, 19.38-31.44), cardiac arrest (aOR, 24.56; 95% CI, 17.84-33.81), and adult respiratory distress syndrome (aOR, 10.72; 95% CI, 8.09-14.20). The failure-to-rescue rate after AFE was 17.0% overall. However, the failure-to-rescue rate exceeded 30% when AFE co-occurred with other severe maternal morbidity indicators: 45.8% for AFE, cardiac arrest, and coagulopathy; 43.2% for AFE, shock, and cardiac rhythm conversion; and 38.6% for AFE, cardiac arrest, coagulopathy, and shock. The failure-to-rescue rate after AFE also exceeded 30% when AFE occurred in the setting of placental pathology: 42.9% for AFE and PAS and 31.3% for AFE and placental abruption.
Conclusions and relevance: This contemporaneous, national-level analysis validated previously known risk factors for AFE and confirmed the dismal outcomes of pregnancy complicated by AFE. The association between PAS and AFE, which was not previously reported, warrants further investigation.