Despite major treatment advances over the past several decades, postpartum hemorrhage (PPH) continues to be a potentially life-threatening obstetric complication and a significant cause of maternal morbidity and mortality, and is also associated with severe medical conditions including hypovolemic shock, multi-organ failure, disseminated intravascular coagulation, acute renal failure, and acute respiratory distress syndrome. The World Health Organization found that PPH affected approximately 2% of all birthing women and accounted for 25% of all global maternal deaths (about 140,000 per year) in 2014 [1]. Statistics Korea [2] has reported that 0.1 patient per 100,000 births dies from massive hemorrhage, accounting for 11% of all maternal deaths in 2014. Thus, in Korea, PPH remains a major complication in the peripartum period, and a common cause of maternal death.

Traditionally, PPH has been defined as a blood loss of ≥500 ml during the peripartum period, and severe PPH as blood loss of ≥1,000 ml. PPH is subdivided into primary PPH (developing within 24 hours of delivery) and secondary PPH (developing between 24 hours and 6 weeks postpartum) [3]. PPH is medically managed using uterotonic agents (e.g., intravenous oxytocin and/or ergometrine, and sublingual misoprostol), and/or transfusion of isotonic crystalloids to replace blood loss. If patients do not respond to uterotonics, thus exhibiting uncontrolled bleeding, nonsurgical interventions including uterine massage, uterine balloon tamponade, uterine artery embolization, and/or surgical interventions, are recommended to treat PPH attributable to uterine atony. Hysterectomy is the treatment of last resort [4].

However, even after such varied treatments, PPH may still cause death or complex problems such as coagulopathy triggered by massive transfusions or amniotic fluidrelated issues. Mortality and morbidity associated with PPH remain high not only in developing countries but also in advanced countries.

Recombinant activated factor VII (rFVIIa; Novoseven, Novo Nordisk, Bagsvaerd, Denmark) is emerging as a novel therapy for the treatment of life- or fertility-threatening PPH unresponsive to standard therapy and may, in some cases, prevent the need for peripartum hysterectomy [5]. However, the World Health Organization considers there is insufficient evidence to strongly recommend use of the drug [4].

In 1998, the effectiveness of rFVIIa was recognized, and rFVIIa was approved for use to control bleeding in patients with hemophilia A or B who expressed inhibitors of coagulation factors. In recent years, hundreds of cases of off-label use of rFVIIa as a hemostatic agent to control severe/massive hemorrhage in a variety of clinical situations have been investigated and the value of rFVIIa has been documented. The outstanding efficacy of off-label rFVIIa in the management of life-threatening obstetric hemorrhage has been recognized by many clinicians, resulting in guidelines for the use of rFVIIa in the management of major obstetric hemorrhage [5]. However, no Korean study has yet evaluated the use of rFVIIa to treat PPH. Thus, we studied the effects of rFVIIa on coagulopathy, transfusion volume required, prognosis, and changes in the severity of PPH in Korean patients.

We treated 66 patients with PPH during the study period; average age was 32.2 ± 5.65 years. There were 15 patients with SOFA scores ≥8. Eight of these 15 patients received injections of rFVIIa whereas seven patients did not (Figure 1). Of the patients treated with rFVIIa, six were primigravida and two multigravida. Of the patients not treated with rFVIIa, four were primigravida and three multigravida. The mean age of the patients included in the analysis was 31.7 ± 7.5 years.

The initial median SOFA score was 10 (interquartile range [IQR], 9.25 to 11.75) in the rFVIIa-use group and 9 (IQR, 8.0 to 10.0) in the non-use group. After 24 hours, the eight patients who received rFVIIa had a median SOFA score of 4 (IQR, 2.25 to 4.75) whereas patients who did not receive rFVIIa had a median score of 5 (IQR, 3.0 to 7.0) (Figure 2). Thus, the median SOFA score did not differ significantly between the two groups. However, the median ΔSOFA value was 6.0 (IQR, 5.3 to 8.5) for the eight patients who recieved rFVIIa and 4.0 (3.0 to 5.0) for the seven patients who did not receive such treatment (P = 0.016) (Table 1, Figure 3).

On comparing the two groups, no significant differences in terms of the transfusion levels of red blood cell concentrate, fresh frozen plasma, or platelet concentrate were evident. Also, there was no significant difference in the duration of either intensive care unit (ICU) stay or hospital stay (Table 1).

Although the incidence of PPH has decreased over the past 40 years, morbidity and mortality associated with obstetric hemorrhage remain high; this is a potentially life-threatening obstetric complication worldwide [1,2]. The standard PPH management consists of initial noninvasive nonsurgical interventions. If necessary, invasive and surgical treatments follow. These include injection of crystalloid and blood components to restore blood volume, uterine massage, uterotonic medications such as oxytocin, uterine compression suturing, vessel ligation, and embolization of the uterine or iliac arteries [4]. In emergency departments, replacing blood volume via massive and aggressive transfusions of crystalloid may trigger serious complications such as coagulopathy, hypothermia, or metabolic acidosis. Also, even the optimal treatments used to arrest obstetric hemorrhage and prevent the need for emergency hysterectomy are only about 50% effective [6].

In recent years, rFVIIa has emerged as a novel therapy for the treatment of major PPH unresponsive to conventional therapies, not only in emergency departments but also in the operative field (Figure 4). Numerous reports of successful off-label use of rFVIIa for the management of major obstetric hemorrhage have appeared. In 2001, the first successful case was reported. That patient received rFVIIa to treat PPH associated with disseminated intravascular coagulation, liver failure, and renal dysfunction [7]. Between 2000 and 2004, a large series of 113 cases were successfully treated in nine European countries; 80% of cases exhibited marked improvement with very few side-effects [8].

In 2006, the European Society of Emergency Medicine, the European Society for Trauma and Emergency Surgery, the European Society of Intensive Care Medicine, the European Society of Anesthesiology, and the European Hematology Association together published revised guidelines for rFVIIa use in the treatment of uncontrolled PPH [9]. Also, an American review group suggested that rFVIIa used to treat PPH was appropriate after failure of clotting factor replacement [10].

In the present study, the amounts of transfused red blood cell concentrate and fresh frozen plasma decreased in most patients following administration of rFVIIa, but the total amounts did not fall significantly. However, we found that the use of rFVIIa to treat severe PPH enhanced SOFA score reduction. Thus, the magnitude of severe PPH can be diminished by rFVIIa use.

The ΔSOFA values correlated with mortality in several studies. Jones et al. [11] analyzed 248 patients with severe sepsis and septic shock and found that initial SOFA score and the ΔSOFA (initial−72 hours) value correlated strongly with mortality. Any increase in the ΔSOFA value was associated with a 35% in-hospital mortality rate, whereas any decrease in the ΔSOFA value was associated with 10% mortality. Ferreira et al. [12] studied 352 patients admitted to a surgical ICU and demonstrated that the mean SOFA score and ΔSOFA value correlated strongly with mortality. Thus, we used the ΔSOFA value to assess the response to treatment, as have other studies [13].

For decades, no clear guidelines for rFVIIa use in situations of massive bleeding have been available. Thus, the doses of rFVIIa used vary among clinicians, ranging from 15 to 120 μg/kg; a single dose usually affords a good result [14]. One report found that a low dose of rFVIIa (i.e., <20 μg/kg) successfully arrested small arterial bleeding [14].

In our present retrospective cohort study, the amount of rFVIIa given was constant (i.e., 250 KIU), thus not based on individual weight, and we did not give repeat injections. Gathering of personal information such as weight in emergency situations is not easy, and the cost of the agent cannot be excluded as a valid consideration. Had we based the rFVIIa levels on individual weight, it would have been possible to evaluate effectiveness more accurately.

The use of rFVIIa is often controversial among clinicians due to the associated complications, especially thrombotic events. In this study, no such complications were encountered. The U.S. Food and Drug Administration warns clinicians of the increased risk of thromboembolic events after rFVIIa administration, particularly in those who have or may have preexisting risk factors. Thus, in practice, clinicians often hesitate to use rFVIIa even in situations of life-threatening obstetric hemorrhage. However, the incidence of adverse events after rFVIIa infusion was only 25 in 100,000 [15]. Bomken et al. [16] reported no thromboembolic event in 15 patients given rFVIIa injections despite the relative prothrombotic states of their pregnancies. In general, many thromboembolic adverse effects developing after rFVIIa injection occurred in patients with preexisting risk factors [17]. Thus, the use of rFVIIa to treat severe PPH is essentially unassociated with thromboembolic events. Clinicians should not be influenced by the controversy; rFVIIa may be considered a safe hemostatic agent for control of major obstetric bleeding.

Our study had several limitations. First, we did not measure fibrinogen levels because such measurement is not routine when patients arrive in the emergency department; the associated laboratory time is long. Thus, we lacked such data. Therefore, the initial and final levels of fibrinogen and platelets did not appear on our data sheets. Second, our patient numbers were small because assessment of rFVIIa off-label use remains incomplete, and the treatment is expensive. Third, we retrospectively reviewed patient data. The absence of randomized controlled studies on rFVIIa use in PPH renders the evidence for such use weak and the data controversial. Several worldwide randomized controlled trials are required to establish clear guidelines for rFVIIa use in cases of major obstetric hemorrhage, and to put the controversy to rest.

In conclusion, we found that the use of rFVIIa to treat severe PPH reduced SOFA score. rFVIIa could be employed as an adjunctive therapy in such cases. Further study is needed, given our small sample size. We view the present work as a pilot study.