Accepted Manuscript Disseminated intravascular coagulation in pregnancy – insights in pathophysiology, diagnosis and management Offer Erez, Salvatore Andrea Mastrolia, Jecko Thachil PII:
S0002-9378(15)00335-X
DOI:
10.1016/j.ajog.2015.03.054
Reference:
YMOB 10342
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 20 January 2015 Revised Date:
26 March 2015
Accepted Date: 29 March 2015
Please cite this article as: Erez O, Mastrolia SA, Thachil J, Disseminated intravascular coagulation in pregnancy – insights in pathophysiology, diagnosis and management, American Journal of Obstetrics and Gynecology (2015), doi: 10.1016/j.ajog.2015.03.054. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Disseminated intravascular coagulation in pregnancy – insights in pathophysiology, diagnosis
Offer EREZ1, Salvatore Andrea MASTROLIA1,2, Jecko THACHIL3
1
RI PT
and management
SC
Department of Obstetrics and Gynecology, Soroka University Medical Center, Faculty of Health
Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel
Department of Obstetrics and Gynecology, Azienda Ospedaliera Universitaria Policlinico di
M AN U
2
Bari, Universita' degli Studi di Bari "Aldo Moro", Bari, Italy 3
Department of Haematology, Manchester Royal Infirmary, Manchester, UK
interest. Corresponding author
EP
Dr. Offer Erez, MD
TE D
Disclosure statement of any potential conflict of interest: The authors report no conflict of
Acting Director Maternal Fetal Medicine Unit
AC C
Department of Obstetrics and Gynecology "B" Soroka University Medical Center Faculty of Health Sciences, Ben Gurion University of the Negev P.O.Box 151, Beer Sheva 84101, Israel Phone +972-8-6400061 [email protected]
1
ACCEPTED MANUSCRIPT
Abstract word count: 132
RI PT
Manuscript word count: 5904
AC C
EP
TE D
M AN U
SC
Condensation: Disseminated intravascular coagulation in pregnancy
2
ACCEPTED MANUSCRIPT
Abstract Disseminated intravascular coagulation (DIC) is a life-threatening situation that can arise from a
RI PT
variety of obstetrical and non-obstetrical causes. Obstetrical DIC has been associated to a series of pregnancy complications including: 1) acute peripartum hemorrhage (uterine atony, cervical and
vaginal
lacerations,
and
uterine
rupture);
2)
placental
abruption;
3)
SC
Preeclampsia/eclampsia/HELLP syndrome; 4) retained stillbirth; 5) septic abortion and intrauterine infection; 6)amniotic fluid embolism; and 7) acute fatty liver of pregnancy. Prompt
M AN U
diagnosis and understanding of the underlying mechanisms of disease leading to this complication in essential for favorable outcome. In the recent years novel diagnostic scores and treatment modalities along with bedside point-of-care tests were developed, and may assist the clinician in the diagnosis and management of DIC. Team work and prompt treatment are
TE D
essential for the successful management of patients with DIC.
EP
Keywords: Acute fatty liver of pregnancy, Endothelial dysfunction, HELLP syndrome,
AC C
Hemorrhage, Score, Trophoblast.
3
ACCEPTED MANUSCRIPT
1. Introduction
Uncontrolled peripartum bleeding, resulting in disseminated intravascular coagulation (DIC), is
RI PT
one of the leading causes for maternal mortality worldwide1. This is in spite of an adaptive physiologic mechanism that generates a physiologic prothrombotic state during gestation2,
3
and the advanced medical and surgical hemostatic capabilities that have evolved during the
SC
past decades for controlling acute obstetrical blood loss.
The rate of DIC during pregnancy differ among cohorts and range from 0.03%4 to 0.35%5. A
M AN U
series of pregnancy complications have been associated with DIC including: 1) acute peripartum hemorrhage (uterine atony, cervical and vaginal lacerations, and uterine rupture); 2) placental abruption; 3) Pre-eclampsia/eclampsia/HELLP syndrome; 4) retained stillbirth; 5) septic abortion and intrauterine infection; 6)amniotic fluid embolism; and 7) acute fatty liver of
TE D
pregnancy1, 6. The proportion of each disorder varies among the different reports. In a cohort5 including 24,693 pregnancies, among those who developed DIC, 49.4% had placental abruption, 29.9% post-partum hemorrhage (PPH), 12.6% severe preeclampsia, and 5.7% had a uterine
EP
rupture. While in a different cohort including 151,678 deliveries, the proportion of these
AC C
complications in those who had DIC was, placental abruption 37%, postpartum hemorrhage or hypovolemia 29%, preeclampsia/HELLP syndrome 14%, acute fatty liver 8%, sepsis 6%, and amniotic fluid embolism 6%4. In most of these pregnancy complications, DIC is associated with adverse maternal outcome including massive blood products transfusion, hysterectomy, and even maternal death7. Therefore, prompt diagnosis and treatment are needed in order to reduce the morbidity and mortality that is associated with DIC.
4
ACCEPTED MANUSCRIPT
In this review we aim to discuss 1) the pathophysiology of DIC focusing on the triad represented by exaggerated activation of coagulation, consumption of coagulopathy and impaired synthesis
RI PT
coagulation as well as anti-coagulation proteins ; 2) the diagnosis of DIC with special attention to the available scores adding prognostic value to the laboratory parameters in patients with this dangerous condition or at risk for its development; and 3) the principles of the treatment of
AC C
EP
TE D
M AN U
SC
DIC, the latter is discussed extensively in the literature6, 8-17.
5
ACCEPTED MANUSCRIPT
2. What is disseminated intravascular coagulation?
Disseminated intravascular coagulation (DIC) represents a life threatening condition which is
RI PT
the endpoint of uncontrolled systemic activation of the hemostatic system, leading to a simultaneous widespread microvascular thrombosis, that can compromise the blood supply to different organs, and may lead to organ failure18. This process is associated with increased
impaired synthesis, leading to uncontrolled bleeding.
SC
degradation of coagulation factors as well as anticoagulation proteins, and followed by their
M AN U
Acute, severe DIC is characterized by diffuse multi-organ bleeding, hemorrhagic necrosis, microthrombi in small blood vessels and thrombi in medium and large blood vessels12. This condition may occur in the setting of sepsis, major trauma, and obstetric calamities. The final scenario is represented by the exhaustion of coagulation/anticoagulation factors and platelets,
TE D
leading to profuse uncontrollable bleeding and often death.
In contrast to the acutely ill patient with complicated severe DIC, other patients may have mild or protracted clinical manifestations of consumption or even subclinical disease manifested by
EP
only laboratory abnormalities19. The clinical picture of sub-acute to chronic DIC is exemplified
AC C
by the chronic hypercoagulability that may accompany malignancy, in particular with mucinproducing adenocarcinomas and acute promyelocytic leukemia. However, currently there are no reports in the literature regarding the occurrence of mild sub-acute DIC in pregnant women. The development of DIC as a result of predisposing conditions can be a life-threatening complication and is considered as one of the leading causes for maternal morbidity and mortality worldwide7. However, it is important to emphasize that DIC is not a disease by itself;
6
ACCEPTED MANUSCRIPT
it is always secondary to an underlying disorder that causes the uncontrolled activation of
AC C
EP
TE D
M AN U
SC
RI PT
coagulation.
7
ACCEPTED MANUSCRIPT
3. What are the mechanisms leading to DIC during pregnancy?
The development of DIC during pregnancy can be either abrupt like in acute abruption or PPH
RI PT
or continuous as can be observed in retained dead fetus. Of interest, obstetric complications such as placental abruption, amniotic fluid embolism, and acute fatty liver of pregnancy are associated with severe early onset DIC that is accompanied by maternal coagulopathy. The DIC
SC
in obstetric hemorrhage activates coagulation and triggers fibrinolysis. Activation of fibrinolysis leads to the production of D-dimers and fibrin-degradation products. These will interfere with
M AN U
platelet function and can impair myometrial contractility20.
Clinical presentation of DIC may be the results of the following mechanisms:
3.1 Endothelial dysfunction and platelet activation: Intact, dysfunctional, or activated cells, as
TE D
well as remnants of cell surfaces, inflammatory mediators and coagulation proteins are all part of an interplay in which uncontrolled activation of coagulation cascade leads to DIC21. Endothelial cells, platelets, but in some cases also leucocytes and cancer cells can participate in
EP
the genesis of the process leading to DIC by releasing pro-inflammatory cytokines, propagating
AC C
the activation of coagulation on their surface or inducing tissue factor (TF) expression on their membrane10, 22-28. A systemic inflammatory response that is associated with markedly increased circulating pro-inflammatory cytokines such as tumor necrosis factor-α, interleukin-1 (IL-1) and interleukin-6 (IL-6) can lead to exaggerated expression of TF by leukocyte and endothelial cells24. This will generate an uncontrolled coagulation response that will eventually deteriorate into DIC. Lastly, the initiation of coagulation leading to thrombin generation in DIC, is mediated by the TF/factor VIIa pathway, also known as the extrinsic coagulation pathway12. The most 8
ACCEPTED MANUSCRIPT
significant source of TF is not completely clear in all situations. Tissue factor may be expressed not only in mononuclear cells in response to pro-inflammatory cytokines (mainly IL-6), but also
RI PT
by vascular endothelial or cancer cells27-29. Despite the potent initiation of coagulation by TF, the activation of coagulation cannot be propagated if the physiological anticoagulant pathways function properly. However, in DIC all major natural anticoagulant pathways [i.e., anti-thrombin
SC
III, protein C system, and TF pathway inhibitor (TFPI)] appear to be impaired30. Plasma concentrations of antithrombin III, the most important inhibitor of thrombin, are markedly
M AN U
reduced during DIC, due to a combination of consumption31, degradation by elastase from activated neutrophils32, and impaired synthesis10. A significant depression of the protein C system may further compromise an adequate regulation of activated coagulation33. This impaired function of the protein C system is caused by a combination of impaired protein
TE D
synthesis, cytokine-mediated down-regulation of endothelial thrombomodulin and a fall in the concentration of the free fraction of protein S (the essential co-factor of protein C), resulting in reduced activation of protein C33. Lastly, there seems to be a misbalance of TFPI function in
EP
relation to the increased TF dependent activation of coagulation34. All these anticoagulant pathways are linked to the endothelium, and it is likely that endothelial
AC C
cell activation and dysfunction are an important component of the imbalance between coagulation and anticoagulation systems. Of interest, experimental and clinical studies indicate that during DIC, the fibrinolytic system is largely suppressed at the time of maximal activation of coagulation10, 11. This inhibition of fibrinolysis is caused by a sustained rise in the plasma concentrations of plasminogen activator inhibitor-1 (PAI-1), the principal inhibitor of the fibrinolytic system. Activation of platelets may also accelerate fibrin formation35. The
9
ACCEPTED MANUSCRIPT
expression of TF in monocytes is markedly stimulated by the presence of platelets and granulocytes in a P-selectin dependent reaction29. This effect may be the result of nuclear factor kappa B (NF-kB) activation induced by binding of activated platelets to neutrophils and
RI PT
mononuclear cells36.
During pregnancy maternal leukocytes are in a higher state of activation than in non-pregnant
SC
women37, and have characteristics akin to sepsis38. However, they are well controlled during pregnancy, and it has been proposed that the trophoblast plays a role in the maintenance of
M AN U
the balanced systemic maternal inflammation during gestation39. Nevertheless, in cases of sepsis due to infectious agent or septic abortion and at least in some of the cases of amniotic fluid embolism40 this equilibrium is disturbed and the mother develops DIC.
TE D
3.2 Trophoblast properties and activation of the coagulation system: During normal gestation the trophoblast has two hemostatic functions: 1) to allow the laminar flow of maternal blood in the intervillous space and prevent it from clotting during that time; and 2) to prevent bleeding
EP
at the maternal fetal interface6. In order to address these contradicting challenges, the syncytiotrophoblast acquires endothelial-cell like properties (Figure 1). As a consequence: 1) In
AC C
human normal placenta syncytiotrophoblast strongly expresses TF, and its activity is higher if compared to human umbilical vein endothelial cells. On the contrary, the trophoblast is able to synthesize Protein C, Protein S and Protein Z as well as a specific inhibitor of the tissue factor pathway known as TFPI-2 (Placental Protein 5)41-43 that will prevent unnecessary activation of the coagulation cascade44-47; 2) The placenta produces PAI-2 in addition to the gradual increase of PAI-1, observed during normal pregnancy and becomes markedly elevated in the third
10
ACCEPTED MANUSCRIPT
trimester in order to prevent fibrinolysis48. These changes are associated with relatively unchanged tissue plasminogen activator (t-PA) concentrations, contributing to a state of reduced clot lysis and a pro-thrombotic bias in the pregnant woman49. This mechanism is (TAFI)50, 51.
RI PT
further mediated through thrombin activatable fibrinolysis inhibitor
The evidence brought herein supports the fact that any condition that disrupts the integrity of
SC
the throphoblast (Figure 2) can lead to a release of a large amount of potent TF that will activate the coagulation cascade and propagate an inflammatory response that can easily
M AN U
become systemic leading to uncontrolled thrombin generation and subsequent development of DIC25, 52. There are several conditions that are associated with DIC in which the current evidence suggests that the systemic maternal response is the result of endothelial activation. The classical one is abruption, especially that with concealed bleeding and fetal demise. These
TE D
patients have a combination of consumption coagulopathy and discharge of thromboplastin (tissue factor) into the maternal circulation10.
Although the DIC developed in patients with placental abruption is regarded as a problem of
EP
consumption coagulopathy, it seems that there is more to it. Meaning, often patients with a retro placental clot have a much lower blood loss than those who developed PPH, yet the DIC
AC C
of in these patients is much more severe. A probable explanation is that this complication is associated to the release of pro-coagulating factors, such as thromboplastin, into the maternal circulation6. In addition, local hypoxia and hypovolemia trigger endothelial response leading to increased expression of vascular endothelial growth factor, which causes an increased endothelial expression of TF53. Evidence in support of this view is brought by Erez at al25 who demonstrated that in women with fetal death, those who had abruption had a higher amniotic
11
ACCEPTED MANUSCRIPT
fluid of TAT complexes. These events result in consumption of coagulating factors, fibrin deposition in microcirculation, and thrombus formation on maternal surface of the placenta at
RI PT
the site of abruption. This is followed by fibrinolysis and release of fibrin degradation products further contributing to the development of DIC. Of interest, if the abruption in concealed or it is severe enough to cause fetal demise, it is at much higher risk for the development of DIC, due
SC
to a continuous release of TF in the maternal circulation. The probable mechanism leading to this observation is similar to that observed in amniotic fluid embolism with systemic release of
M AN U
TF that leads to systemic activation of coagulation and subsequent DIC. This view is supported by the experiment reported by Schneider54 which demonstrated that the intravenous injection of placental extracts into mice leads to the death of the animal through DIC, which can be prevented by the administration of heparin54. The author identified thromboplastin as the
TE D
causative agent through its effect on the clotting time and its chemical properties, and measured its activity by the one-stage prothrombin-time method54.
6, 18, 23, 55-58
as a
EP
3.3 Hemorrhage: Acute obstetrical bleeding is being considered by many4,
leading cause for DIC. This form of consumption coagulopathy is classically related to PPH as a
AC C
result of uterine atony, retained placenta or membranes, uterine rupture, placenta accreta, or severe cervical or vaginal lacerations5. In all of these cases, the mother is losing a large volume of blood and coagulation factors in a short time interval, and these patients are usually hemodynamically compromised. Currently, there is a debate whether this form of consumption coagulopathy is truly DIC or just a massive blood loss that depletes the patient's coagulation factors and can lead to death due to exanguination9. However, massive maternal bleeding may
12
ACCEPTED MANUSCRIPT
not be that straightforward as pure loss of coagulation factors. During the time of parturition and postpartum period there is substantial activation of coagulation cascade and generation of thrombin as a result of the release of TF to the maternal circulation following the separation of
RI PT
the membranes and the placenta59, 60. Thus, these women already have increased thrombin generation61 and indeed are regarded as high risk patients for the development of deep vein
SC
thrombosis during the puerperium62. The evidence brought herein, that parturients with PPH have a higher activation of coagulation cascade even above the physiologic threshold, suggest
M AN U
that the clinicians who treat these patients must regard them as a high risk group for DIC, even though the fundamental pathology is a rapid and massive loss of blood as well as coagulation factors52. Therefore, patients with PPH need to be treated promptly, pharmacologically and/or surgically, and by blood products as well as volume expanders to sustain the maternal
TE D
circulation and maybe to prevent the subsequent development of DIC.
3.4 Disruption of liver function
EP
3.4.1 Acute fatty liver of pregnancy: Acute fatty liver of pregnancy is a rare, (an estimated incidence between 6 and 14 per 100,000 pregnancies63-65) but potentially fatal complication of
AC C
pregnancy. It is characterized by fatty microvascular infiltration of hepatocytes with progressive loss of liver function66, without alteration of the overall structure of the liver. Women who develop this complication have abnormal renal function67 and DIC68. The mechanisms in which DIC develops in this complication is a combination of reduced liver production of fibrinogen as well as coagulation proteins, and hemorrhage. Evidence in support of this view is presented by Nelson et al69, who studied 51 women with acute fatty liver of pregnancy. Their hemostatic
13
ACCEPTED MANUSCRIPT
condition was classified according to the International Society of Thrombosis and Haemostasis DIC score70, 71 and 80% of these women had unequivocal DIC defined as composite score of 5 or
RI PT
greater. The authors studied the hepatic and hemostatic function of these patients including fibrinogen, fibrin–fibrinogen split products, coagulation studies, and cholesterol. Those who developed DIC had abnormally low plasma fibrinogen concentrations that persisted for the first several days after delivery along with only mild to moderately elevated fibrin degradation
SC
products69. At the same time, there was also evidence for continuing increased procoagulant
M AN U
consumption caused by ongoing DIC provided by the modestly elevated levels of fibrin degradation products in the face of depressed plasma fibrinogen concentrations. This observation was in contrast to that of patients with abruption in whom the fibrinogen concentration recovered into normal range several hours after the acute event. Collectively, the
TE D
continuous low fibrinogen concentration and abnormal function of the coagulation cascade is the result of the liver dysfunction associated with acute fatty liver of pregnancy, leading to a
EP
lower production of coagulation factors, anticoagulation proteins and fibrinogen by the liver.
3.4.2 HELLP syndrome: this condition is an additional cause for DIC in obstetric patients that
AC C
may involve the liver. The relationship between acute fatty liver of pregnancy and HELLP syndrome has not been clearly established. There are obviously common clinical and biological features between these two entities68. Indeed, some authors66, 72 have suggested that acute fatty liver of pregnancy and HELLP syndrome are the two faces of the same coin. However, others73,
74
found that a difference in liver histopathology (fatty microvascular infiltration of
hepatocytes vs. fibrin deposition or hemorrhage in the periportal areas75, 76) makes an overlap
14
ACCEPTED MANUSCRIPT
between these two entities not possible. One of the major differences between acute fatty liver of pregnancy and HELLP syndrome is the prevalence of DIC. In a study68 by Vigil-De Gracia, DIC
RI PT
was present in >70% of patients with acute fatty liver of pregnancy and 4 g/l
SC
had a negative predictive value of 79%. For this reason, importance once again should be given to a decreasing fibrinogen level rather than an absolute value with a threshold value of 1.5 g/dl
M AN U
or even higher recommended for replacement86. Fibrin degradation product measurements also have its problems in pregnancy since D-dimers are already raised even before pathological states set in. Repeated measurements showing increasing values may be helpful. Point-of-care testing using devices like thromboelastography (TEG; Haemonetics, Braintree MA,
TE D
USA), or thromboelastometry (ROTEM; TEM GmBH, Munich, Germany) is useful in the obstetrical coagulopathic disorders to achieve rapid results and decide intervention. Normal ranges have been published for women at the time of delivery compared to the standard
EP
ranges90. Collins et al85 reported the ROTEM Fibtem A5 assay as a very useful marker for monitoring haemostasis in this setting. In addition, Sharma et al91 have proposed a
AC C
thromboelastographic score, studying 21 patients classified following the International Society for Haemostasis and Thrombosis (ISTH) score. They demonstrated that parameters arising from this technique may reach a sensitivity of 95.2%, specificity of 81.0%, with the highest receiver operating characteristic area under the curve of all parameters of 0.957 for identifying overt DIC. The use of thromboelastometry has been further evaluated in patients with DIC related to
18
ACCEPTED MANUSCRIPT
severe sepsis showing it can be a valuable tool in assessing whole blood coagulation capacity in patients with severe sepsis with and without overt DIC92.
RI PT
Data on thromboelastography and thromboelastometry in pregnant women are limited, especially during peripartum period and in women with PPH, so more research in this field is needed. Moreover, preliminary data are encouraging, since these tests may be able to detect
SC
early alteration of coagulation pathways and hyperfibrinolysis93 allowing, in combination with the other diagnostic and prognostic means, like DIC scores, an adequate surveillance and,
M AN U
eventually, a prompt intervention in obstetric not yet severely affected patients94.
The acutely bleeding patient does not need any score evaluation but prompt infusion of blood products according to preexisting protocols. However, in many cases, DIC develops gradually
TE D
and through different underlying mechanisms that are described in the manuscript. In the latter group, using such a scoring system can alert the clinician that his patient is deteriorating and needs further attention and treatment. In addition, the introduction of such scoring system into
EP
clinical work will help to validate the diagnosis and create a common language between clinician and researchers that will assist in the promotion of the understanding and
AC C
management of DIC during pregnancy. To address the task of facilitating the diagnosis of DIC, in those conditions that do not represent emergencies in which only a clinical diagnosis can be achieved, the use of scoring systems has been introduced. The first DIC score has been recommended by the ISTH in 200170 showing a correlation between key clinical observations and outcomes. Using the same parameters, the Japanese Association for Acute Medicine (JAAM) published an additional score in 200595, offering a good predictive value for the
19
ACCEPTED MANUSCRIPT
diagnosis of DIC and the identification of critically ill non-pregnant patients. These scores can be used not only as a diagnostic but also as prognostic tool. Thus, in the non-pregnant state a DIC score is important in the diagnosis of patients with DIC and carries a diagnostic and prognostic
RI PT
value71, 96-98.
Since the physiologic hemostatic changes occurring in pregnancy affect the application of these scores to gestation, an adjusted score for the pregnant state was needed. Based on this
SC
consideration Erez et al5, have recently constructed a pregnancy modified DIC score by using
M AN U
only three components of the ISTH DIC score (platelet count, fibrinogen concentrations and the PT difference) with an area the curve of 0.975 (p
S0002-9378(15)00335-X
DOI:
10.1016/j.ajog.2015.03.054
Reference:
YMOB 10342
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 20 January 2015 Revised Date:
26 March 2015
Accepted Date: 29 March 2015
Please cite this article as: Erez O, Mastrolia SA, Thachil J, Disseminated intravascular coagulation in pregnancy – insights in pathophysiology, diagnosis and management, American Journal of Obstetrics and Gynecology (2015), doi: 10.1016/j.ajog.2015.03.054. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Disseminated intravascular coagulation in pregnancy – insights in pathophysiology, diagnosis
Offer EREZ1, Salvatore Andrea MASTROLIA1,2, Jecko THACHIL3
1
RI PT
and management
SC
Department of Obstetrics and Gynecology, Soroka University Medical Center, Faculty of Health
Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel
Department of Obstetrics and Gynecology, Azienda Ospedaliera Universitaria Policlinico di
M AN U
2
Bari, Universita' degli Studi di Bari "Aldo Moro", Bari, Italy 3
Department of Haematology, Manchester Royal Infirmary, Manchester, UK
interest. Corresponding author
EP
Dr. Offer Erez, MD
TE D
Disclosure statement of any potential conflict of interest: The authors report no conflict of
Acting Director Maternal Fetal Medicine Unit
AC C
Department of Obstetrics and Gynecology "B" Soroka University Medical Center Faculty of Health Sciences, Ben Gurion University of the Negev P.O.Box 151, Beer Sheva 84101, Israel Phone +972-8-6400061 [email protected]
1
ACCEPTED MANUSCRIPT
Abstract word count: 132
RI PT
Manuscript word count: 5904
AC C
EP
TE D
M AN U
SC
Condensation: Disseminated intravascular coagulation in pregnancy
2
ACCEPTED MANUSCRIPT
Abstract Disseminated intravascular coagulation (DIC) is a life-threatening situation that can arise from a
RI PT
variety of obstetrical and non-obstetrical causes. Obstetrical DIC has been associated to a series of pregnancy complications including: 1) acute peripartum hemorrhage (uterine atony, cervical and
vaginal
lacerations,
and
uterine
rupture);
2)
placental
abruption;
3)
SC
Preeclampsia/eclampsia/HELLP syndrome; 4) retained stillbirth; 5) septic abortion and intrauterine infection; 6)amniotic fluid embolism; and 7) acute fatty liver of pregnancy. Prompt
M AN U
diagnosis and understanding of the underlying mechanisms of disease leading to this complication in essential for favorable outcome. In the recent years novel diagnostic scores and treatment modalities along with bedside point-of-care tests were developed, and may assist the clinician in the diagnosis and management of DIC. Team work and prompt treatment are
TE D
essential for the successful management of patients with DIC.
EP
Keywords: Acute fatty liver of pregnancy, Endothelial dysfunction, HELLP syndrome,
AC C
Hemorrhage, Score, Trophoblast.
3
ACCEPTED MANUSCRIPT
1. Introduction
Uncontrolled peripartum bleeding, resulting in disseminated intravascular coagulation (DIC), is
RI PT
one of the leading causes for maternal mortality worldwide1. This is in spite of an adaptive physiologic mechanism that generates a physiologic prothrombotic state during gestation2,
3
and the advanced medical and surgical hemostatic capabilities that have evolved during the
SC
past decades for controlling acute obstetrical blood loss.
The rate of DIC during pregnancy differ among cohorts and range from 0.03%4 to 0.35%5. A
M AN U
series of pregnancy complications have been associated with DIC including: 1) acute peripartum hemorrhage (uterine atony, cervical and vaginal lacerations, and uterine rupture); 2) placental abruption; 3) Pre-eclampsia/eclampsia/HELLP syndrome; 4) retained stillbirth; 5) septic abortion and intrauterine infection; 6)amniotic fluid embolism; and 7) acute fatty liver of
TE D
pregnancy1, 6. The proportion of each disorder varies among the different reports. In a cohort5 including 24,693 pregnancies, among those who developed DIC, 49.4% had placental abruption, 29.9% post-partum hemorrhage (PPH), 12.6% severe preeclampsia, and 5.7% had a uterine
EP
rupture. While in a different cohort including 151,678 deliveries, the proportion of these
AC C
complications in those who had DIC was, placental abruption 37%, postpartum hemorrhage or hypovolemia 29%, preeclampsia/HELLP syndrome 14%, acute fatty liver 8%, sepsis 6%, and amniotic fluid embolism 6%4. In most of these pregnancy complications, DIC is associated with adverse maternal outcome including massive blood products transfusion, hysterectomy, and even maternal death7. Therefore, prompt diagnosis and treatment are needed in order to reduce the morbidity and mortality that is associated with DIC.
4
ACCEPTED MANUSCRIPT
In this review we aim to discuss 1) the pathophysiology of DIC focusing on the triad represented by exaggerated activation of coagulation, consumption of coagulopathy and impaired synthesis
RI PT
coagulation as well as anti-coagulation proteins ; 2) the diagnosis of DIC with special attention to the available scores adding prognostic value to the laboratory parameters in patients with this dangerous condition or at risk for its development; and 3) the principles of the treatment of
AC C
EP
TE D
M AN U
SC
DIC, the latter is discussed extensively in the literature6, 8-17.
5
ACCEPTED MANUSCRIPT
2. What is disseminated intravascular coagulation?
Disseminated intravascular coagulation (DIC) represents a life threatening condition which is
RI PT
the endpoint of uncontrolled systemic activation of the hemostatic system, leading to a simultaneous widespread microvascular thrombosis, that can compromise the blood supply to different organs, and may lead to organ failure18. This process is associated with increased
impaired synthesis, leading to uncontrolled bleeding.
SC
degradation of coagulation factors as well as anticoagulation proteins, and followed by their
M AN U
Acute, severe DIC is characterized by diffuse multi-organ bleeding, hemorrhagic necrosis, microthrombi in small blood vessels and thrombi in medium and large blood vessels12. This condition may occur in the setting of sepsis, major trauma, and obstetric calamities. The final scenario is represented by the exhaustion of coagulation/anticoagulation factors and platelets,
TE D
leading to profuse uncontrollable bleeding and often death.
In contrast to the acutely ill patient with complicated severe DIC, other patients may have mild or protracted clinical manifestations of consumption or even subclinical disease manifested by
EP
only laboratory abnormalities19. The clinical picture of sub-acute to chronic DIC is exemplified
AC C
by the chronic hypercoagulability that may accompany malignancy, in particular with mucinproducing adenocarcinomas and acute promyelocytic leukemia. However, currently there are no reports in the literature regarding the occurrence of mild sub-acute DIC in pregnant women. The development of DIC as a result of predisposing conditions can be a life-threatening complication and is considered as one of the leading causes for maternal morbidity and mortality worldwide7. However, it is important to emphasize that DIC is not a disease by itself;
6
ACCEPTED MANUSCRIPT
it is always secondary to an underlying disorder that causes the uncontrolled activation of
AC C
EP
TE D
M AN U
SC
RI PT
coagulation.
7
ACCEPTED MANUSCRIPT
3. What are the mechanisms leading to DIC during pregnancy?
The development of DIC during pregnancy can be either abrupt like in acute abruption or PPH
RI PT
or continuous as can be observed in retained dead fetus. Of interest, obstetric complications such as placental abruption, amniotic fluid embolism, and acute fatty liver of pregnancy are associated with severe early onset DIC that is accompanied by maternal coagulopathy. The DIC
SC
in obstetric hemorrhage activates coagulation and triggers fibrinolysis. Activation of fibrinolysis leads to the production of D-dimers and fibrin-degradation products. These will interfere with
M AN U
platelet function and can impair myometrial contractility20.
Clinical presentation of DIC may be the results of the following mechanisms:
3.1 Endothelial dysfunction and platelet activation: Intact, dysfunctional, or activated cells, as
TE D
well as remnants of cell surfaces, inflammatory mediators and coagulation proteins are all part of an interplay in which uncontrolled activation of coagulation cascade leads to DIC21. Endothelial cells, platelets, but in some cases also leucocytes and cancer cells can participate in
EP
the genesis of the process leading to DIC by releasing pro-inflammatory cytokines, propagating
AC C
the activation of coagulation on their surface or inducing tissue factor (TF) expression on their membrane10, 22-28. A systemic inflammatory response that is associated with markedly increased circulating pro-inflammatory cytokines such as tumor necrosis factor-α, interleukin-1 (IL-1) and interleukin-6 (IL-6) can lead to exaggerated expression of TF by leukocyte and endothelial cells24. This will generate an uncontrolled coagulation response that will eventually deteriorate into DIC. Lastly, the initiation of coagulation leading to thrombin generation in DIC, is mediated by the TF/factor VIIa pathway, also known as the extrinsic coagulation pathway12. The most 8
ACCEPTED MANUSCRIPT
significant source of TF is not completely clear in all situations. Tissue factor may be expressed not only in mononuclear cells in response to pro-inflammatory cytokines (mainly IL-6), but also
RI PT
by vascular endothelial or cancer cells27-29. Despite the potent initiation of coagulation by TF, the activation of coagulation cannot be propagated if the physiological anticoagulant pathways function properly. However, in DIC all major natural anticoagulant pathways [i.e., anti-thrombin
SC
III, protein C system, and TF pathway inhibitor (TFPI)] appear to be impaired30. Plasma concentrations of antithrombin III, the most important inhibitor of thrombin, are markedly
M AN U
reduced during DIC, due to a combination of consumption31, degradation by elastase from activated neutrophils32, and impaired synthesis10. A significant depression of the protein C system may further compromise an adequate regulation of activated coagulation33. This impaired function of the protein C system is caused by a combination of impaired protein
TE D
synthesis, cytokine-mediated down-regulation of endothelial thrombomodulin and a fall in the concentration of the free fraction of protein S (the essential co-factor of protein C), resulting in reduced activation of protein C33. Lastly, there seems to be a misbalance of TFPI function in
EP
relation to the increased TF dependent activation of coagulation34. All these anticoagulant pathways are linked to the endothelium, and it is likely that endothelial
AC C
cell activation and dysfunction are an important component of the imbalance between coagulation and anticoagulation systems. Of interest, experimental and clinical studies indicate that during DIC, the fibrinolytic system is largely suppressed at the time of maximal activation of coagulation10, 11. This inhibition of fibrinolysis is caused by a sustained rise in the plasma concentrations of plasminogen activator inhibitor-1 (PAI-1), the principal inhibitor of the fibrinolytic system. Activation of platelets may also accelerate fibrin formation35. The
9
ACCEPTED MANUSCRIPT
expression of TF in monocytes is markedly stimulated by the presence of platelets and granulocytes in a P-selectin dependent reaction29. This effect may be the result of nuclear factor kappa B (NF-kB) activation induced by binding of activated platelets to neutrophils and
RI PT
mononuclear cells36.
During pregnancy maternal leukocytes are in a higher state of activation than in non-pregnant
SC
women37, and have characteristics akin to sepsis38. However, they are well controlled during pregnancy, and it has been proposed that the trophoblast plays a role in the maintenance of
M AN U
the balanced systemic maternal inflammation during gestation39. Nevertheless, in cases of sepsis due to infectious agent or septic abortion and at least in some of the cases of amniotic fluid embolism40 this equilibrium is disturbed and the mother develops DIC.
TE D
3.2 Trophoblast properties and activation of the coagulation system: During normal gestation the trophoblast has two hemostatic functions: 1) to allow the laminar flow of maternal blood in the intervillous space and prevent it from clotting during that time; and 2) to prevent bleeding
EP
at the maternal fetal interface6. In order to address these contradicting challenges, the syncytiotrophoblast acquires endothelial-cell like properties (Figure 1). As a consequence: 1) In
AC C
human normal placenta syncytiotrophoblast strongly expresses TF, and its activity is higher if compared to human umbilical vein endothelial cells. On the contrary, the trophoblast is able to synthesize Protein C, Protein S and Protein Z as well as a specific inhibitor of the tissue factor pathway known as TFPI-2 (Placental Protein 5)41-43 that will prevent unnecessary activation of the coagulation cascade44-47; 2) The placenta produces PAI-2 in addition to the gradual increase of PAI-1, observed during normal pregnancy and becomes markedly elevated in the third
10
ACCEPTED MANUSCRIPT
trimester in order to prevent fibrinolysis48. These changes are associated with relatively unchanged tissue plasminogen activator (t-PA) concentrations, contributing to a state of reduced clot lysis and a pro-thrombotic bias in the pregnant woman49. This mechanism is (TAFI)50, 51.
RI PT
further mediated through thrombin activatable fibrinolysis inhibitor
The evidence brought herein supports the fact that any condition that disrupts the integrity of
SC
the throphoblast (Figure 2) can lead to a release of a large amount of potent TF that will activate the coagulation cascade and propagate an inflammatory response that can easily
M AN U
become systemic leading to uncontrolled thrombin generation and subsequent development of DIC25, 52. There are several conditions that are associated with DIC in which the current evidence suggests that the systemic maternal response is the result of endothelial activation. The classical one is abruption, especially that with concealed bleeding and fetal demise. These
TE D
patients have a combination of consumption coagulopathy and discharge of thromboplastin (tissue factor) into the maternal circulation10.
Although the DIC developed in patients with placental abruption is regarded as a problem of
EP
consumption coagulopathy, it seems that there is more to it. Meaning, often patients with a retro placental clot have a much lower blood loss than those who developed PPH, yet the DIC
AC C
of in these patients is much more severe. A probable explanation is that this complication is associated to the release of pro-coagulating factors, such as thromboplastin, into the maternal circulation6. In addition, local hypoxia and hypovolemia trigger endothelial response leading to increased expression of vascular endothelial growth factor, which causes an increased endothelial expression of TF53. Evidence in support of this view is brought by Erez at al25 who demonstrated that in women with fetal death, those who had abruption had a higher amniotic
11
ACCEPTED MANUSCRIPT
fluid of TAT complexes. These events result in consumption of coagulating factors, fibrin deposition in microcirculation, and thrombus formation on maternal surface of the placenta at
RI PT
the site of abruption. This is followed by fibrinolysis and release of fibrin degradation products further contributing to the development of DIC. Of interest, if the abruption in concealed or it is severe enough to cause fetal demise, it is at much higher risk for the development of DIC, due
SC
to a continuous release of TF in the maternal circulation. The probable mechanism leading to this observation is similar to that observed in amniotic fluid embolism with systemic release of
M AN U
TF that leads to systemic activation of coagulation and subsequent DIC. This view is supported by the experiment reported by Schneider54 which demonstrated that the intravenous injection of placental extracts into mice leads to the death of the animal through DIC, which can be prevented by the administration of heparin54. The author identified thromboplastin as the
TE D
causative agent through its effect on the clotting time and its chemical properties, and measured its activity by the one-stage prothrombin-time method54.
6, 18, 23, 55-58
as a
EP
3.3 Hemorrhage: Acute obstetrical bleeding is being considered by many4,
leading cause for DIC. This form of consumption coagulopathy is classically related to PPH as a
AC C
result of uterine atony, retained placenta or membranes, uterine rupture, placenta accreta, or severe cervical or vaginal lacerations5. In all of these cases, the mother is losing a large volume of blood and coagulation factors in a short time interval, and these patients are usually hemodynamically compromised. Currently, there is a debate whether this form of consumption coagulopathy is truly DIC or just a massive blood loss that depletes the patient's coagulation factors and can lead to death due to exanguination9. However, massive maternal bleeding may
12
ACCEPTED MANUSCRIPT
not be that straightforward as pure loss of coagulation factors. During the time of parturition and postpartum period there is substantial activation of coagulation cascade and generation of thrombin as a result of the release of TF to the maternal circulation following the separation of
RI PT
the membranes and the placenta59, 60. Thus, these women already have increased thrombin generation61 and indeed are regarded as high risk patients for the development of deep vein
SC
thrombosis during the puerperium62. The evidence brought herein, that parturients with PPH have a higher activation of coagulation cascade even above the physiologic threshold, suggest
M AN U
that the clinicians who treat these patients must regard them as a high risk group for DIC, even though the fundamental pathology is a rapid and massive loss of blood as well as coagulation factors52. Therefore, patients with PPH need to be treated promptly, pharmacologically and/or surgically, and by blood products as well as volume expanders to sustain the maternal
TE D
circulation and maybe to prevent the subsequent development of DIC.
3.4 Disruption of liver function
EP
3.4.1 Acute fatty liver of pregnancy: Acute fatty liver of pregnancy is a rare, (an estimated incidence between 6 and 14 per 100,000 pregnancies63-65) but potentially fatal complication of
AC C
pregnancy. It is characterized by fatty microvascular infiltration of hepatocytes with progressive loss of liver function66, without alteration of the overall structure of the liver. Women who develop this complication have abnormal renal function67 and DIC68. The mechanisms in which DIC develops in this complication is a combination of reduced liver production of fibrinogen as well as coagulation proteins, and hemorrhage. Evidence in support of this view is presented by Nelson et al69, who studied 51 women with acute fatty liver of pregnancy. Their hemostatic
13
ACCEPTED MANUSCRIPT
condition was classified according to the International Society of Thrombosis and Haemostasis DIC score70, 71 and 80% of these women had unequivocal DIC defined as composite score of 5 or
RI PT
greater. The authors studied the hepatic and hemostatic function of these patients including fibrinogen, fibrin–fibrinogen split products, coagulation studies, and cholesterol. Those who developed DIC had abnormally low plasma fibrinogen concentrations that persisted for the first several days after delivery along with only mild to moderately elevated fibrin degradation
SC
products69. At the same time, there was also evidence for continuing increased procoagulant
M AN U
consumption caused by ongoing DIC provided by the modestly elevated levels of fibrin degradation products in the face of depressed plasma fibrinogen concentrations. This observation was in contrast to that of patients with abruption in whom the fibrinogen concentration recovered into normal range several hours after the acute event. Collectively, the
TE D
continuous low fibrinogen concentration and abnormal function of the coagulation cascade is the result of the liver dysfunction associated with acute fatty liver of pregnancy, leading to a
EP
lower production of coagulation factors, anticoagulation proteins and fibrinogen by the liver.
3.4.2 HELLP syndrome: this condition is an additional cause for DIC in obstetric patients that
AC C
may involve the liver. The relationship between acute fatty liver of pregnancy and HELLP syndrome has not been clearly established. There are obviously common clinical and biological features between these two entities68. Indeed, some authors66, 72 have suggested that acute fatty liver of pregnancy and HELLP syndrome are the two faces of the same coin. However, others73,
74
found that a difference in liver histopathology (fatty microvascular infiltration of
hepatocytes vs. fibrin deposition or hemorrhage in the periportal areas75, 76) makes an overlap
14
ACCEPTED MANUSCRIPT
between these two entities not possible. One of the major differences between acute fatty liver of pregnancy and HELLP syndrome is the prevalence of DIC. In a study68 by Vigil-De Gracia, DIC
RI PT
was present in >70% of patients with acute fatty liver of pregnancy and 4 g/l
SC
had a negative predictive value of 79%. For this reason, importance once again should be given to a decreasing fibrinogen level rather than an absolute value with a threshold value of 1.5 g/dl
M AN U
or even higher recommended for replacement86. Fibrin degradation product measurements also have its problems in pregnancy since D-dimers are already raised even before pathological states set in. Repeated measurements showing increasing values may be helpful. Point-of-care testing using devices like thromboelastography (TEG; Haemonetics, Braintree MA,
TE D
USA), or thromboelastometry (ROTEM; TEM GmBH, Munich, Germany) is useful in the obstetrical coagulopathic disorders to achieve rapid results and decide intervention. Normal ranges have been published for women at the time of delivery compared to the standard
EP
ranges90. Collins et al85 reported the ROTEM Fibtem A5 assay as a very useful marker for monitoring haemostasis in this setting. In addition, Sharma et al91 have proposed a
AC C
thromboelastographic score, studying 21 patients classified following the International Society for Haemostasis and Thrombosis (ISTH) score. They demonstrated that parameters arising from this technique may reach a sensitivity of 95.2%, specificity of 81.0%, with the highest receiver operating characteristic area under the curve of all parameters of 0.957 for identifying overt DIC. The use of thromboelastometry has been further evaluated in patients with DIC related to
18
ACCEPTED MANUSCRIPT
severe sepsis showing it can be a valuable tool in assessing whole blood coagulation capacity in patients with severe sepsis with and without overt DIC92.
RI PT
Data on thromboelastography and thromboelastometry in pregnant women are limited, especially during peripartum period and in women with PPH, so more research in this field is needed. Moreover, preliminary data are encouraging, since these tests may be able to detect
SC
early alteration of coagulation pathways and hyperfibrinolysis93 allowing, in combination with the other diagnostic and prognostic means, like DIC scores, an adequate surveillance and,
M AN U
eventually, a prompt intervention in obstetric not yet severely affected patients94.
The acutely bleeding patient does not need any score evaluation but prompt infusion of blood products according to preexisting protocols. However, in many cases, DIC develops gradually
TE D
and through different underlying mechanisms that are described in the manuscript. In the latter group, using such a scoring system can alert the clinician that his patient is deteriorating and needs further attention and treatment. In addition, the introduction of such scoring system into
EP
clinical work will help to validate the diagnosis and create a common language between clinician and researchers that will assist in the promotion of the understanding and
AC C
management of DIC during pregnancy. To address the task of facilitating the diagnosis of DIC, in those conditions that do not represent emergencies in which only a clinical diagnosis can be achieved, the use of scoring systems has been introduced. The first DIC score has been recommended by the ISTH in 200170 showing a correlation between key clinical observations and outcomes. Using the same parameters, the Japanese Association for Acute Medicine (JAAM) published an additional score in 200595, offering a good predictive value for the
19
ACCEPTED MANUSCRIPT
diagnosis of DIC and the identification of critically ill non-pregnant patients. These scores can be used not only as a diagnostic but also as prognostic tool. Thus, in the non-pregnant state a DIC score is important in the diagnosis of patients with DIC and carries a diagnostic and prognostic
RI PT
value71, 96-98.
Since the physiologic hemostatic changes occurring in pregnancy affect the application of these scores to gestation, an adjusted score for the pregnant state was needed. Based on this
SC
consideration Erez et al5, have recently constructed a pregnancy modified DIC score by using
M AN U
only three components of the ISTH DIC score (platelet count, fibrinogen concentrations and the PT difference) with an area the curve of 0.975 (p
