Clinical Context

Postpartum hemorrhage is the leading cause of maternal mortality globally and a significant cause of US maternal morbidity and mortality. Risk factors include clinical factors causing uterine atony, uterine overdistention, cesarean delivery, uterine fibroids, and advanced age.

Many of these risk factors appear to be increasing in the population. Recent US population trends in postpartum hemorrhage risk and outcomes in the setting of recent clinical management changes are poorly characterized.

Study Synopsis and Perspective

The rate of postpartum hemorrhage for hospital deliveries in the United States increased significantly over the course of a 20-year period, according to data from more than 76 million delivery hospitalizations from the National Inpatient Sample.

Postpartum hemorrhage remains the leading cause of maternal morbidity and mortality worldwide, and many clinical- and patient-level risk factors appear to be on the rise, write Chiara M. Corbetta-Rastelli, MD, from the University of California, San Francisco, and colleagues.

Although practice changes have been introduced to reduce postpartum hemorrhage, recent trends in postpartum hemorrhage risk and outcomes in the context of such changes as hemorrhage safety bundles have not been examined, they said.

In a study published in Obstetrics & Gynecology, the researchers reviewed data from hospitalizations for females aged 15 to 54 years for deliveries between 2000 and 2019, using the National Inpatient Sample.^[1] They used a regression analysis to estimate average annual percentage changes (AAPC). Their objectives were to characterize trends and to assess the association between risk factors and the occurrence of postpartum hemorrhage and related interventions. Demographics, clinical factors, and hospital characteristics were mainly similar between the group of patients with postpartum hemorrhage and those with no postpartum hemorrhage.

Approximately 3% (2.3 million) of 76.7 million hospitalizations for delivery were complicated by postpartum hemorrhage during the study period, and the annual rate increased from 2.7% to 4.3%.

Overall, 21.4% of individuals with delivery hospitalizations complicated by postpartum hemorrhage had 1 postpartum risk factor and 1.4% had 2 or more risk factors. The number of individuals with at least 1 risk factor for postpartum hemorrhage increased significantly, from 18.6% to 26.9%, during the study period, with an annual percentage change of 1.9%.

Compared with deliveries in individuals without risk factors, individuals with 1 risk factor had slightly higher odds of postpartum hemorrhage (odds ratio, 1.14), but those with 2 or more risk factors were more than twice as likely to experience postpartum hemorrhage as those with no risk factors (odds ratio, 2.31).

The researchers also examined the association of specific risk factors and interventions related to hemorrhage--notably, blood transfusion and peripartum hysterectomy. Blood transfusions in individuals with postpartum hemorrhage increased from 5.4% to 16.7% between 2000 and 2011 (AAPC, 10.2%) and then decreased from 16.7% to 12.6% from 2011 to 2019 (AAPC, −3.9%).

Peripartum hysterectomy in the study population increased from 1.4% to 2.4% from 2000 to 2009 (AAPC 5.0%), remained steady from 2009 to 2016, and then decreased from 2.1% to 0.9% from 2016 to 2019 (AAPC, −27%).

Other risk factors associated with postpartum hemorrhage itself and with blood transfusion and hysterectomy in the setting of postpartum hemorrhage included prior cesarean delivery with placenta previa or accreta, placenta previa without prior cesarean delivery, and antepartum hemorrhage or placental abruption, the researchers note.

“In addition to placental abnormalities, risk factors such as preeclampsia with severe features, polyhydramnios, and uterine leiomyomas demonstrated the highest rates of increase in our data,” they write in their discussion. These trends may lead to continuing increases in postpartum hemorrhage risk, which was not fully explained by the increase in risk factors seen in the current study, the researchers said.

The study findings were limited by several factors, including the use of billing codes that could lead to misclassification of diagnoses, as well as possible differences in the definition and coding for postpartum hemorrhage among hospitals, the researchers note. Other limitations were the exclusion of cases of readmission for postpartum hemorrhage and lack of clinical details involving use of medications or nonoperative interventions, they said.

Notably, the study finding of stable to decreasing peripartum hysterectomy rates in hospitalized patients with postpartum hemorrhage conflicts with another recent study showing an increase in peripartum hysterectomy from 2009 to 2020, but this difference may reflect changes in billing, indications for hysterectomy, or study modeling, they note.^[2]

The current study was strengthened by the use of a large database to analyze population trends, a contemporary study period, and the inclusion of meaningful outcomes such as peripartum hysterectomy.

The shift in blood transfusion and peripartum hysterectomy may reflect the implementation of protocols to promote early intervention and identification of postpartum hemorrhage, they conclude.

“Hemorrhage remains a leading cause of maternal mortality in the United States, and blood transfusion is the most common severe maternal morbidity,” Catherine M. Albright, MD, associate professor of maternal-fetal medicine at the University of Washington, Seattle, said in an interview. “It is important to understand the current state, especially given that many hospitals have implemented policies and procedures to better identify and treat postpartum hemorrhage,” she said.

Dr Albright added, “I was pleased to see that they did not just look at a diagnosis of postpartum hemorrhage but rather also looked at complications arising from postpartum hemorrhage, such as blood transfusion or hysterectomy.”

Postpartum hemorrhage is often a clinical diagnosis that uses estimated blood loss, which is a notoriously inaccurate measure, said Dr Albright. “Additionally, the definitions of postpartum hemorrhage, as well as the [International Classification of Diseases] codes, changed during the time period of the study,” she noted. “These factors all could lead to both underreporting and overreporting of the true incidence of postpartum hemorrhage. Blood transfusion and hysterectomy are more objective outcomes and demonstrate true morbidity,” she said.

“Most of the risk factors that are listed in the article are not modifiable during that pregnancy,” said Dr Albright. For example, a history of a prior cesarean or having a twin pregnancy is not something that can be changed, she said. “Many of the other risk factors or associated clinical factors, such as obesity, chronic hypertension, and pregestational diabetes, are modifiable, but before pregnancy. Universal and easy access to primary medical care prior to and between pregnancies may help to mitigate some of these factors,” she noted.

Looking ahead, “It would be helpful to ensure that these types of data are available at the state and hospital level; this will allow for local evaluation of programs that are in place to reduce postpartum hemorrhage risk and improve identification and treatment,” Dr Albright said.

The study received no outside funding. Dr Corbetta-Rastelli and Dr Albright have disclosed no relevant financial relationships.

Obstet Gynecol. 2023;141(1):152-161

Clinical Implications

• Postpartum hemorrhage and related risk factors increased over the course of 2 decades, but blood transfusions and hysterectomies recently decreased.
• Increase in risk factors may contribute to, but not fully explain, continuing increases in postpartum hemorrhage risk.
• Implications for the Health Care Team: The care team should consider the use of standardized protocols or safety bundles, for early recognition and more timely medical management of modifiable risk factors.

Earn Credit