Clinical Management of Postpartum Hemorrhage

© Springer International Publishing Switzerland 2016
Marco Ranucci and Paolo Simioni (eds.)Point-of-Care Tests for Severe Hemorrhage10.1007/978-3-319-24795-3_9

9. Clinical Management of Postpartum Hemorrhage

Vanessa Agostini , Maria Pia Rainaldi2, Maria Grazia Frigo3, Massimo Micaglio4 and Agostino Brizzi5

Transfusion Medicine Cesena/Forlì and Blood Establishment, Clinical Pathology Department-Cesena Azienda U.S.L Romagna, Romagna, Italy

Department of Obstetrics and Gynecology, Sant’Orsola Malpighi University Hospital, University of Bologna, Bologna, Italy

Obstetric Anesthesia, Fatebenefratelli Hospital Isola Tiberina Rome, Rome, Italy

Department of Anesthesia and Intensive Care, Careggi University Hospital, Florence, Italy

Department of Anesthesia and Intensive Care, Santa Maria Hospital, Bari, Italy



Vanessa Agostini

Postpartum hemorrhage (PPH) is the leading cause of severe maternal morbidity and mortality worldwide.

Its management is complex and requires timely interventions, skilled personnel, human, technological and financial resources, and well-organized maternity services. A treatment failure or near misses could be the results of misdiagnosis or a delay in diagnosis of the underestimation of the bleeding rate and of a lack of adequate education and training together with a deficiency in the organization.

This chapter focuses the attention on the hemostatic management of severe obstetric hemorrhage.

9.1 Definitions and Epidemiology

A satisfactory definition of PPH does not exist, but traditionally, primary PPH is defined as a blood loss of 500 mL or more from the genital tract within 24 h from the birth of a baby, and it can be minor (500–1000 mL), moderate (1000–2000 mL), or severe (more than 2000 mL). Massive PPH has been defined as a bleeding >2500 mL and is associated with significant morbidity such as hysterectomy and need for intensive care admission (http://​www.​healthcareimprov​ementscotland.​org). The amount of blood loss is very difficult to clinically assess and is usually underestimated [1]. Other definitions of PPH are the following: a decline in hemoglobin of 4 g/dL or more, transfusion of at least 4 units of red blood cells, or the need for embolization or operative intervention.

A recent systematic review [2] has estimated a global prevalence of PPH to be 10.8 % and severe PPH to be 2.8 %.

It is well known that there is a regional variation of postpartum hemorrhage (≥500 mL) ranging from 7.2 % in Oceania to 25.1 % in Africa, while the prevalence of a blood loss equal or above 1000 mL is higher in Africa (5.1 %) and lower in Asia (1.9 %).

In developed countries such as Europe and North America, the prevalence is around 13 %, being higher for multiple pregnancies (32.4 % compared with 10.6 % for singletons) and for first-time mothers (12.9 % compared with 10.0 % for women in subsequent pregnancies). Furthermore there are some evidences of an increased incidence in PPH in Canada, the UK, Australia, and the USA between 2001 and 2006 related to a rise in obesity.

On the contrary it is possible that the lower prevalence of PPH in Asian women is related to regional differences in genetics or underlying risk factors such as a lower incidence of obesity in Asia.

9.2 The Pathogenesis of PHH

There are several causes of PPH, classified according to their underlying pathophysiology: tone (uterine atony), tissue (placenta accreta, percreta, increta, previa), trauma (laceration of cervix, vagina), and thrombin (congenital coagulation disorders). Uterine atony is the leading cause of massive obstetric hemorrhage, difficult to be predicted.

9.2.1 Pregnancy-Induced Coagulation Changes

It is important to highlight that at term the hemostatic system is unbalanced toward a prothrombotic state, due to an increase of procoagulant factors VII, VIII, IX, X, XII, von Willebrand, and fibrinogen. In particular the fibrinogen level is markedly increased, reaching 4–6 g/L compared with the nonpregnant normal range of 2–4 g/L. Prothrombin (FII) and factor V remain unchanged, while factor XI is somewhat reduced [3, 4]. On the other hand, there is a decrease of the natural anticoagulant protein S level. These changes are responsible for the shorter prothrombin time (PT) and activated partial thromboplastin time (aPTT) and for the increase of thromboelastometric parameters such as maximum amplitude (MA) and maximum clot firmness (MCF) [5].

Plasma fibrinolytic activity is reduced during pregnancy, remains low during labor, and increases at the time of delivery due to the reduction of the fibrinolytic inhibitors (PAI-1 and 2) and to the high levels of plasminogen tissue factor (t-PA).

Five different studies with different entry criteria demonstrated that low fibrinogen levels in the early stages of PPH have similar predictive values. In particular, a fibrinogen level <2 g/L is associated with progression to severe PPH.

Charbit and associates [6] showed that in women at the time of a second-line uterotonic for atony, a fibrinogen level <2 g/L had a positive predictive value (PPV) of 100 % for progression to severe PPH, while a level >4 g/L had a negative predictive value of 79 %.

Collins and associates [7] recently investigated the utility of FIBTEM A5 and Clauss fibrinogen as predictors of progression of PPH in a consecutive cohort of 356 women in a 1-year period (2012–2013), all experiencing a bleeding >1000 mL. Entry criteria were a bleeding of 1000–1499 plus cesarean section or uterine atony or placental abruption or placenta previa or microvascular bleeding or cardiovascular instability or any bleeding >1500 mL. Blood samples were collected before any administration of prohemostatic drugs or blood component therapy.

The primary outcome was to evaluate the utility of FIBTEM A5 and/or Clauss fibrinogen as predictors of progression to massive hemorrhage (>2500 mL). Other outcomes were transfusion of <4 or ≥4 units of red blood cells, transfusion <8 or ≥8 of any blood product (red blood cells, fresh frozen plasma, or platelets), need for an invasive procedure, bleeding duration, and length of stay in the high-dependency unit.

The positive predictive value for progression to any transfusion of a fibrinogen <2 g/L or FIBTEM A5 <10 mm was 75 and 71 %, respectively; these values reached 100 % in women with ongoing bleeding at study entry. For progression to massive PPH (>2500 mL) in the final multivariate analysis, only FIBTEM A5 was an independent predictor.

The median fibrinogen level of women transfused with ≥4 units RBCs was 2.6 g/L compared with 3.9 g/L for those not transfused.

During PPH, coagulopathies may develop secondary to the dilution related to volume replacement and to the loss and consumption of coagulation factors. All these factors lead to a fall in fibrinogen levels and platelet count and affect thrombin generation.

9.3 Coagulation Monitoring

To date there are mainly two different strategies to monitor coagulation changes during PPH: the first is based on standard laboratory tests (PT/aPTT and Clauss fibrinogen) and the second on point-of-care coagulation testing devices such as thromboelastometry/thromboelastography.

Despite the hemostatic derangement, standard coagulation tests (PT/aPTT) remain normal until the blood loss reaches 4000–5000 mL, while the fibrinogen levels fall earlier than the other coagulation factors [8]. Recent data show that a low Clauss fibrinogen (<2 g/L) and a reduced maximum amplitude at the ROTEM device are accurate biomarkers for the progression to severe PPH, and the recent ESA guidelines [9] recommend thromboelastometric measurements in order to identify the contribution of fibrinogen to clot strength.

Conventional coagulation tests typically require 45–60 min to be available; they are not adequate to diagnose hyperfibrinolysis, and their results may not influence the initial resuscitation although they are a useful baseline for later management of blood component therapy.

Viscoelastic tests such as rotational thromboelastometry or thromboelastography are dynamic tests which give information on the whole coagulation process from the initiation of clot formation through fibrinolysis. The major advantages of point-of-care testing are that the clinicians can rapidly identify if the bleeding has a coagulopathic or a surgical origin. Furthermore there is good evidence that the ROTEM FIBTEM A5 assay can be used as a surrogate for Clauss fibrinogen during PPH [7, 10, 11].

In a prospective observational study, Houissoud and associates [11] compared 51 women with PPH to a control group and demonstrated that in ROTEM, FIBTEM clotting amplitudes were lower in the PPH group in comparison with the control group. Furthermore a strong correlation between fibrinogen levels and the FIBTEM parameters was present in both groups. The Clauss fibrinogen threshold of 2 g/L corresponded to a FIBTEM A5 of 6 mm and to a FIBTEM A15 of 8 mm with a sensitivity of 100 % both and a specificity of 87 and 84 %, respectively. It is important to consider that thromboelastometry normal ranges at the time of delivery differ from the nonpregnant normal range. Armstrong and associates [12] studied ROTEM parameters from 60 healthy pregnant women presenting for elective cesarean section and 60 matched nonpregnant female controls presenting for elective surgery and demonstrated that pregnant women had significant lower INTEM CT, CFT, and EXTEM CFT (p < 0.001) and higher MCF in INTEM, EXTEM, and FIBTEM (p < 0.001) in comparison to the control nonpregnant group.

De Lange et al. [5] in a Dutch multicenter study reported ROTEM reference values after having collected blood samples from 161 women at two different times: time 1 during labor between 6 and 10 cm cervical dilatation or before elective cesarean section and time 2 within 1 h of delivery of the placenta. They found a shorter CT and CFT in EXTEM and INTEM and larger A10, A20, and MCF in INTEM and FIBTEM in pregnant women in comparison with reference values in nonpregnant adults. Parameters before and after labor were not different, and there was a strong correlation between ROTEM FIBTEM and fibrinogen concentration.

All these data are extremely important because of the increasing use of viscoelastic tests in postpartum hemorrhage management and in order to define specific triggers for hemostatic therapy in this setting.

9.4 The Hemostatic Management of PPH

Rapid diagnosis and early treatment of PPH are critical for a favorable patient outcome. It is well known that the underestimation of the speed and extent of hemorrhage, delay in blood transfusion, lack of treatment algorithm, lack of adequate education and training, poor communication, and deficiencies in organization [13] are responsible for failures in treatment.

The hemostatic management of PPH is challenging, and current guidelines rely on experience from the nonpregnant population and expert opinion.

Three different approaches are used to date for the management of PPH: a laboratory-driven approach, a formula-driven one, and, more recently, a goal-directed strategy based on viscoelastic coagulation monitoring.

9.4.1 Laboratory-Driven Strategy

This approach derived from the past guidelines [1416] recommends to start red blood cell (RBC) transfusion as soon as possible together with warm crystalloid and colloid solutions for the volume replacement.

Fresh frozen plasma (FFP) should be used when the results of PT/aPTT are available and >1.5 × normal at the standard dose of 12–15 mL/kg or a total of 1 l every 6 units of RBC. Platelet concentrates (PLT) should be transfused if platelet count <50 × 109/L and cryoprecipitated (2 pools) if fibrinogen level <1 g/L with the goal to keep:

  • Hemoglobin >8 g/dl

  • Platelet count >75 × 109/L

  • PT/aPTT <1.5 × normal control

  • Fibrinogen >1 g/L

This kind of approach is now considered unsatisfactory especially in case of severe/massive PPH and in case of hemodynamically unstable patients, when there is no time to wait for the result of standard laboratory test.

9.4.2 The Formula-Driven Approach

In order to save time, most centers have recently adopted the strategy used for the management of trauma-induced coagulopathy based on the delivery of predefined “shock packs” (the so-called formula-driven approach).

This strategy, derived from trauma experiences, consists in the adoption of formulaic protocols based on the release of blood component therapy in fix ratio 1:1 RBC/FFP or 1:1:1 RBC/FFP/PLT, adding cryoprecipitate later on during resuscitation. A number of retrospective studies, in the last decade, have reported a correlation between earlier and higher FFP to RBC transfusion ratios and improved survival and outcomes. In comparison to traumatology data, fewer studies have been conducted in obstetric setting.

In 2007 Burtelow and associates [17] described the implementation of a massive transfusion protocol for life-threatening primary postpartum hemorrhage consisting in the release of 6 units of group 0 D-negative RBC, 4 units of FFP, and 1 apheresis unit of platelets (fixed ratio 6:4:1). This fixed pack allowed the blood bank to quickly release blood component therapy during the acute phase of resuscitation and the clinicians to anticipate and probably prevent dilutional coagulopathy. After the first pack (6:4:1), the protocol consisted in the release of other FFP, platelets, and cryoprecipitate units on the basis of the standard coagulation laboratory results. The same group in 2011 [18] analyzed the impact of their massive transfusion protocol (fix pack 6:4:1) on blood inventory and the wastage rates of blood components.

The rationale for a fixed ratio approach is to recapitulate whole blood and to maintain thrombin generation and fibrinogen concentration by the replacement of coagulation factors as soon as possible.

However recent data from trauma research demonstrated the inability of this strategy to maintain good levels of fibrinogen.

Moreover the major disadvantage of a formulaic unmonitored approach is the risk of an overtreatment of women with a normal coagulation profile at admission and a dilution effect on fibrinogen level, if it is considered that plasma derived from donors contains approximately 2 g/L of fibrinogen.

Early empirical transfusion of FFP might be justified in cases:

  • Of consumption (e.g., amniotic fluid embolism)

  • Of expected very large volumes of blood loss (placenta abnormalities, e.g., placenta accreta)

  • Of hemodynamically instability

9.4.3 The Goal-Directed Approach

This strategy was started in German-speaking countries since 20 years ago, and it can be considered a tactical approach to the problem. It is based on the rapid differential diagnosis of coagulopathies through the use of viscoelastic coagulation monitoring devices (ROTEM, TEG) with the possibility to step by step treat the coagulation abnormality without using an empirical or formulaic approach, avoiding unnecessary blood component transfusions and using prohemostatic drugs such as antifibrinolytic agents, fibrinogen concentrate, and prothrombin complex concentrates.

Different algorithms were recently published from different groups.

Girard and associates in 2014 [19] proposed an algorithm from a German-speaking countries’ (Austria, Germany, and Switzerland) working group of obstetricians, anesthesiologists, and hemostasis specialists.

The algorithm is based on four steps:

Jul 20, 2017 | Posted by in Uncategorized | Comments Off on Clinical Management of Postpartum Hemorrhage

Full access? Get Clinical Tree

Get Clinical Tree app for offline access