According to the World Health Statistics, as of 2017, approximately 295,000 maternal deaths occur worldwide every year, and the global maternal mortality rate is as high as 2.11/100,000 [1]. The maternal mortality rate in high-income countries is 11/100,000.

Postpartum hemorrhage (PPH) is the leading cause of maternal death in the world. In addition, severe postpartum hemorrhage (sPPH) is defined as blood loss $\ge$1000 mL within 24 hours after delivery in the ninth edition of Chinese textbook of Obstetrics and Gynecology, which is considered a life-threatening condition. Therefore, the prediction of sPHH occurrence during pregnancy is essential to ensure the safety of pregnant women. In this study, we have analyzed the high-risk factors that contribute to sPPH.

The education level, body mass index (BMI) before pregnancy, BMI before delivery, single pregnancy, multiple pregnancy, previous history of postpartum hemorrhage, previous history of cesarean section, birth weight, hemoglobin before delivery (g/L), platelet count before delivery ($\times$10${}^9$), prepartum coagulation function prothrombin time (PT), prepartum activated partial thromboplastin time (APTT), prepartum fibrinogen (FIB) (g/L), pre-transfusion hemoglobin (g/L), pre-transfusion platelet ($\times$10${}^9$), pre-transfusion coagulation function PT, pre-transfusion APTT, pre-transfusion TT, pretransfusion FIB (g/L), marginal placenta previa, placenta accreta, central placenta previa, low lying placenta, IVF-ET pregnancy and antepartum hemorrhage were risk factors for severe PPH according to t-test or rank sum test (Table 1).

The BMI before pregnancy, past history of postpartum hemorrhage, birth weight prepartum APTT, prepartum FIB (g/L), pre-transfusion hemoglobin (g/L), pre-transfusion platelet ($\times$10${}^9$), pre-transfusion coagulation function prothrombin time (PT), marginal placenta previa, placenta accreta, central placenta previa, IVF-ET pregnancy and antepartum hemorrhage were risk factors for severe PPH according to multivariate analysis (Table 2).

Hence, many clinical research studies have focused on assessing obstetric complications. In this study, we found that most deaths occurred within 24 h after the delivery of the baby; a compensated state can rapidly transition to decompensated state and is easily overlooked. PPH can be reduced through the identification of high-risk patients, early surgical preparations, and continued vigilance. PPH can be caused by uterine atony, birth canal injury, placental factors, and coagulation dysfunction. These four major causes are interconnected and mutually causal [2]. About 70%–80% of PPH is due to abnormal uterine tension (uterine atony). It is considered the primary cause of postpartum hemorrhage [3]. In this study, we have demonstrated that there are statistically significant differences if certain factors such as previous history of cesarean section, marginal placenta previa, placenta accreta, central placenta previa, and low-lying placenta are compared between the study and the control groups. Also, marginal placenta previa, placenta accreta and central placenta previa are the separate risk factors for sPPH.

Li et al. [4] found that placenta previa could increase the incidence of PPH by 6 to 20 times. They also reported that the incidence of PPH in patients with placenta previa can vary from 22.06% to 42.09%. The incidence of PPH in patients with placenta previa is as high as 38.20% [5], and the incidence of PPH in central placenta previa is 58% [6]. We also confirmed that placental disease is an independent risk factor that contributes significantly (foremost factor) to sPPH. Miscarriages, intrauterine operations and application of IVF-ET are also contributing factors to placental diseases. The larger the area of the internal cervical os covered by the placenta, the higher the risk of PPH. Central placenta previa is more likely to cause sPPH than marginal placenta previa or low lying placenta. The main reason for the formation of placenta previa is primary decidual dysplasia or an endometrial defect, mostly caused by trauma, resulting in a lack of the decidua basalis and adequate blood supply. The placenta covers the lower segment of the uterus and the internal os of the cervix [7]. The placenta has abundant blood vessels, but the muscle layer at this lower segment is weak, and the placenta has abundant blood vessels with resulting weak uterine contractility. When the placenta is stripped during labor, the blood sinuses on the stripped surface are open, which then induces PPH. In addition, the wound closure is delayed, and the amniotic fluid potentially enters the maternal blood vessels inducing acute disseminated intravascular coagulation, thus, affecting coagulation function and leading to excessive bleeding. Li et al. [4] reported that placenta accreta increases the risk of PPH by 7 to 11 times. Placenta accreta is the condition when the villi invade the myometrium and prevents the placenta from separating from the uterine wall. Insufficient uterine contractions or prolonged opening of blood sinuses during delivery may lead to difficulty in blood coagulation on the adhesion surface, which can also contribute to PPH. Therefore, preventing placental disease during pregnancy, especially central placenta previa with placenta accreta, and improving the existing standard treatment for placenta accreta, specifically the surgical procedures, are keys to preventing sPPH.

DiMarco et al. [8] found that pregnant patients with a medical history of multiple abortions have a significantly higher incidence of postpartum complications. The main reason could be that the uterine contractility is significantly reduced after multiple abortions, and the placenta is more likely to form abnormally in subsequent pregnancy, increasing the risk of PPH. We report that a history of multiple miscarriages is related to sPPH [9]. Therefore, medical personnel should educate young women about the proper use of contraception and its advantages. High-risk factors that cause placental disease should be considered in order to develop preventive measures and to avoid pregnancy complications, especially in women with short stature or those with insufficient weight gain during pregnancy.

Cui et al. [10] confirmed that the level of fibrinogen positively correlates with gestational age. A high level of fibrinogen can decrease the PT and partially activated PT which may reduce the occurrence of PPH. Fibrinogen varies from 2.4 to 4 g/L in adults, but it ranges from 3.4 to 5.4 g/L in the third trimester of pregnancy. The amount of maternal plasma fibrinogen has important predictive significance for PPH. van Dijk et al. [11] has found that the platelet count before childbirth is also closely related to the occurrence of PPH with the risk of PPH being high when the platelet count is low. Our results corroborate this finding.

This study demonstrates that a previous history of PPH, IVF-ET pregnancy, prepartum APTT, prepartum FIB (g/L), pre-transfusion hemoglobin (g/L), pre-transfusion platelet ($\times$10${}^9$) and pre-transfusion coagulation function prothrombin time (PT) are independent risk factors for sPPH. A higher level of hemoglobin (g/L) and prepartum hematocrit (%) can reduce the risk of sPPH. Therefore, detection and timely treatment of anemia during pregnancy are important for the prevention of sPPH [12]. Screening and therapeutical management of coagulation disorders during pregnancy and immediately before childbirth should also be considered a preventive measure for sPPH [13]. Diseases that may cause coagulation disorders include obstetric antiphospholipid syndrome (OAPS), thrombophilia, systemic lupus erythematosus, Sjögren’s syndrome, lupus nephritis, and other autoimmune diseases. Patients exhibiting any of these entities are prone to sPPH. One of the prevention and treatment plans that can effectively control the rate of bleeding is to immediately start the blood transfusion process. It diminishes the probability of the rapid onset of sPPH.

The system, infrastructure, technology, and trained personnel are all present. It is critical to strictly abide by the core system to follow the diagnosis and treatment specifications and provide team simulation training and individual skill training to healthcare professionals. We summarize it as the “5 early” principles, indicating early prediction, early warning, early prevention, early planning and early treatment are important to manage PPH. The etiology for sPPH is diverse and complicated in many pregnant patients during childbirth. For women of childbearing age, especially those exhibiting high-risk factors, evaluating the risk of PPH before pregnancy, providing special health care during pregnancy and delivery, taking preventive measures in advance, and managing the bleeding by medications are the keys to reducing sPPH [14]. A comprehensive treatment strategy for sPPH can effectively reduce the morbidity and mortality for pregnant patients.

The availability of these data is restricted due to privacy or ethical considerations. The data supporting the findings of this study can be obtained upon request from the corresponding author.

GZ contributed to the design of the study and conception of content. YZhang and JZ contributed to the study concepts, study design and data analysis. KG contributed to the literature research, data acquisition and manuscript preparation. YG contributed to the data curation and analysis. YZhu contributed to the literature research and manuscript preparation. All authors have read and approved the final version of the manuscript. All authors contributed to editorial changes in the manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was approved by the institutional review board of Shijiazhuang Fourth Hospital (approval number 20190067) in accordance with the October 2013 edition of the Declaration of Helsink. All methods were carried out in accordance with relevant guidelines and regulations. Written informed consent was obtained from all individual patients included in the study.

We would like to express our gratitude to all those who helped us during the writing of this manuscript. Thanks to all the peer reviewers for their opinions and suggestions.

This work was supported by Hebei Provincial Department of Science and Technology Special Project for People’s Livelihood Science and Technology (NO.202030701180541). This research was supported by the Key Research and Development Program of Hebei Province (No.20377779D).

The authors declare no conflict of interest.