Fig. 4.1
Coagulation cascade
Levels of FXIII increase initially and then gradually decrease, reaching 50 % of the normal non-pregnant levels by delivery [14]. Prothrombin and FV do not change significantly during pregnancy [10].
Antithrombin is a serine protease inhibitor which is synthesized in the liver, and its main anticoagulant effect is through the inhibition of thrombin (IIa) and FXa. There are controversial reports about changes in antithrombin levels throughout pregnancy but concentrations generally show little change (Fig. 4.2) [11]. Thrombin generation is also regulated by protein C and its cofactor protein S. These are both vitamin K-dependent glycoproteins and are synthesized mainly in the liver. Protein C is activated by thrombin bound to thrombomodulin (TM), a glycoprotein receptor found on the surface of endothelial cells. This is then enhanced by a second glycoprotein receptor, the endothelial protein C receptor (EPCR). Both these receptors are also expressed by placental trophoblasts. Protein C levels appear to remain within normal levels in pregnancy [11, 12].
Fig. 4.2
Natural anticoagulants
Protein S exists in plasma in two forms. Approximately 60 % is bound to complement 4b-binding protein (C4BP) and is inactive, the remaining 40 % circulates in the active form as free protein S. There is a progressive decrease in protein S levels in the plasma during pregnancy [15]. As early as 6–11 weeks, levels of total and free protein S are below the non-pregnant range for women who are not on oral contraceptives [8]. It is, therefore, very difficult to make a diagnosis of protein S deficiency in pregnancy as levels can fall to those found in heterozygous protein S deficiency.
Activated protein C (APC) resistance is found in association with the Factor V Leiden mutation or as an acquired state in association with antiphospholipid antibodies or cancer. APC resistance is also found during pregnancy. At term 45 % of women have an APC sensitivity ratio below the 5th percentile of the reference range for non-pregnant women of a similar age [16]. The reduction in APC ratio is directly related to its value outside of pregnancy, being most pronounced in those with the highest APC ratio [11, 12]. About 50 % of healthy women develop APC resistance, which reaches its lowest value in the second trimester with little further change [11]. This behavior of the classical APC resistance test has been called “acquired” APC resistance [11, 12]. Asymptomatic pregnant carriers of the Factor V Leiden mutation do not exhibit more pronounced coagulation activation than non-carriers [17]. However, higher levels of d-dimer are found in women with the Factor V Leiden mutation, indicating an increase in fibrin formation and activation of the fibrinolytic system. While it is likely that an increase in FVIII and FIX levels and a decrease in free protein S contribute to the acquired APC resistance seen in pregnancy, the correlation between these variables has not been consistently reported [8]. In a longitudinal study on healthy pregnant women [18], no correlation was found between the total change in the classical APC ratio and the total changes in FVIII, fibrinogen, or protein S.
The principal fibrinolytic enzyme is plasmin, which circulates as the inactive zymogen, plasminogen. The activation of plasmin is mediated by two types of plasminogen activator: tissue-type and urokinase-type (uPA). Tissue plasminogen activator (tPA) is released into blood by the endothelium. It activates plasminogen to plasmin when both fibrinogen and tPA bind to the fibrin clot. Plasmin is inhibited directly by the plasmin inhibitor (α2-antiplasmin) and α2-macroglobulin and indirectly (through the inhibition of tPA) by plasminogen activator inhibitor-1 (PAI-1) produced by endothelial cells and platelets (Fig. 4.3). The activation of plasmin is reduced by thrombin-activatable fibrinolysis inhibitor (TAFI) [1].
Fig. 4.3
Fibrinolytic pathway
Plasma fibrinolytic activity is reduced in pregnancy, remains low during labor, and returns to normal shortly after delivery [10]. T-PA activity decreases during pregnancy. This is due to the increase in PAI-1 and to the production of plasminogen activator inhibitor-2 (PAI-2) by the placenta. At 35 weeks PAI-1 levels are fivefold higher than in the 12th week of pregnancy. PAI-1 levels normalize by 5 weeks post-partum [11, 12].
Levels of PAI-2 become detectable in plasma only in pregnant women. Villous cells are the source of PAI-2 and, therefore, changes in the amount of placental tissue may influence its plasma levels. A positive correlation between placental birth weight and PAI-2 levels has been found [11, 12]. Despite high levels of PAI-1 and PAI-2 in pregnancy, a highly significant positive correlation between gestational age and d-dimer has been observed [19, 20]. The increase in d-dimer throughout pregnancy limits its use in exclusion of venous thromboembolism (VTE) in pregnant women with clinical suspicion.
Levels of TAFI have been reported to remain stable over months during pregnancy and to be correlated only with age in women [11, 12]. No correlation has been found between TAFI and d-dimer levels (Table 4.1) [11].
Clotting factors | Change |
|---|---|
Platelet Count | Decreased |
FII, FV | Unchanged |
Fibrinogen, FVII, FVIII, VWF, FIX, FX, FXII | Increased |
FXI | Unchanged/decreased |
FXIII | Increased/decreased |
Antithrombin | Unchanged |
Protein C | Unchanged/increased |
Protein S | Decreased |
Heparin cofactor II | Increased |
F 1+2, TAT, d-dimer | Increased |
t-PA | Decreased |
ELT, PAI, TAFI | Increased |
Microparticles
Microparticles (MP) are membrane vesicles that are shed from various cellular surfaces. There are two mechanisms that can result in MP formation. These are cell activation and apoptosis. MP are associated with inflammatory and thrombotic complications. When exposed to cytokines, such as interleukin-1 and tumor necrosis factor, endothelial cells produce MP. Circulating platelet MP is an indicator of platelet activation. Normal pregnancy is characterized by an increase in endothelial and platelet MP compared to healthy, non-pregnant women. However, the role of MP in gestational vascular complications remains controversial [11, 12].
Mechanism of Thrombosis in Assisted Conception
The number of women undergoing assisted conception is increasing worldwide. In the U.K., more than 60,000 cycles of in vitro fertilization (IVF) are carried out annually, leading to more than 14,000 births [21]. The incidence of thrombosis in women undergoing assisted conception is small and similar to that of pregnancy-associated thrombosis [22]. This has been reported at a rate of 0.08–0.11 % of all treatment cycles for venous events, although the rate of arterial events is likely to be lower [23].
Ovarian stimulation, designed to increase the number of eggs and embryos available, is used in the vast majority of these cycles. The most significant short-term complication associated with ovarian stimulation is ovarian hyperstimulation syndrome (OHSS, Chap. 3). Although cases of spontaneous OHSS have been reported, generally OHSS is an iatrogenic complication of assisted reproductive techniques [24]. OHSS is a systemic disease which results from vasoactive products which are released from hyperstimulated ovaries. It is characterised by increased vascular permeability which leads to third-space fluid accumulation and intravascular dehydration. Early and late OHSS are different entities which differ in severity and predisposing factors. Late-onset (after 9 days from HCG trigger) OHSS is often associated with successful embryo implantation resulting in pregnancy and is more severe than early-onset OHSS. OHSS is further classified into mild, moderate, severe, and critical, depending on the abdominal distension, amount of ascites, ovarian size, and involvement of other organs [21]. Mild OHSS affects up to 33 % of IVF cycles but only 3–8 % are complicated my moderate or severe OHSS [25]. It is important to remember that OHSS is a dynamic condition and the level of severity can change over time.
OHSS is associated with an increased risk of both venous and arterial thromboses, and these are a serious and life-threatening complication of OHSS. Venous events in this situation are more common (75 %) than arterial events (25 %) [24]. Arterial thromboses are most likely due to thromboembolic events, commonly at an intracerebral location and are almost always concurrent with the development of symptoms of OHSS [24, 26]. Chan reported 35 cases of arterial thrombosis; 60 % of these were cerebrovascular accidents, 17 % involved the extremities, and 11 % were myocardial infarcts [23]. OHSS was found in 90 % of all the cases of arterial thrombosis. This is in contrast to venous events, which can occur several weeks after resolution of OHSS (with a reported range of 2 days to 11 weeks) [23]. Venous thrombosis can present as classical deep-vein thrombosis (DVT), but is often observed in unusual sites such as the upper limb and neck veins [24]. Chan reported 61 cases of venous thrombosis; 80 % of these venous events were reported in the neck and upper extremities [23]. Thrombosis can occur in women with OHSS who are not pregnant, and therefore OHSS represents an independent risk factor for the development of thrombosis.
Several factors contribute to the thrombotic risk in OHSS: haemoconcentration, leucocytosis, thrombocytosis, accompanied by alterations in the coagulation and fibrinolytic systems. Haemoconcentration leads to an elevated blood viscosity, a slower blood stream, and vascular stasis [24].
A number of studies have highlighted prothrombotic changes during the IVF process. These include increases in levels of VWF, FVIII, and fibrinogen, along with decreased levels of proteins C, S and antithrombin [23]. In a study of 12 women undergoing ovarian stimulation, significantly increased levels of fibrinogen and a reduction in antithrombin was reported [27]. This study also showed a significant reduction in fibrinolysis as measured by the euglobulin clot lysis time (ELT). A further case control study [28] showed that tissue factor, d-dimer, thrombin-antithrombin complexes (TAT), prothrombin fragment 1+2 (F1+2), plasmin-antiplasmin complexes, and VWF antigen levels in plasma were significantly higher in the group with severe OHSS than case-controls (women undergoing IVF who did not develop OHSS) and healthy controls (healthy age-matched women). In contrast, tissue factor pathway inhibitor (TFPI) levels were significantly lower.
Global measures of haemostasis using the thromboelastogram have suggested hypercoagulable changes, although the parameters remain within normal limits [29]. These changes appear to be more pronounced in women who develop OHSS [30]. The analysis of longitudinal studies has concluded that activation of both the coagulation system and fibrinolysis occurs in women who undergo ovarian hyperstimulation and that this is greatly exaggerated in OHSS [23].
Development of OHSS has been shown to be associated with inherited and acquired thrombophilias. In one study, positive markers for thrombophilia (such as low levels of antithrombin, protein S and protein C, positive antiphopholipid antibodies, and the Factor V Leiden mutation) found in 85 % of patients with severe OHSS and only 27 % of women without OHSS following ovarian stimulation [31]. It therefore appears that thrombophilia is not only associated with an increased risk of thrombosis but also an increased risk of developing OHSS. Thrombophilia in assisted reproduction is discussed further in the next section of this chapter.
As well as changes to the coagulation system during OHSS, there are other hypotheses to account for the increased risk of venous thrombosis and the unusual sites at presentation. Bauersachs et al. postulated that ascitic fluid, high in oestrogen, drains into the thoracic duct [32]. This fluid then drains into the left subclavian vein, resulting in a high concentration of oestrogen in that local area, leading to thrombosis in the neck veins. Another hypothesis is that rudimentary brachial cysts in the neck fill with fluid during OHSS and cause mechanical obstruction at the base of the jugular and subclavian veins. This then leads to upper extremity thrombosis [33].
Overall, the risk of developing thrombosis during assisted conception is small and similar to that of normal pregnancy. However, development of OHSS increases this risk, particularly in cases of severe OHSS. It is recommended that all hospitalised women with severe OHSS receive pharmacological thromboprophylaxis and wear full-length graduated compression stockings. There are no guidelines on the duration of prophylaxis when OHSS is associated with pregnancy. However, the risk of thrombosis appears to persist into the first trimester of pregnancy and consideration should be given to the risks and benefits of thromboprophylaxis until the end of the first trimester, or even longer [25]. Thromboprophylaxis is discussed in more detail later in the chapter.
Inherited and Acquired Thrombophilia in Assisted Conception
Routine testing for inherited thrombophilia prior to assisted conception is not recommended in current U.K. guidelines [25, 34], but is a contentious area. It may be of value in guiding treatment in those women who have a personal or family history of thrombosis, or if they have had three unexplained miscarriages, placental abruption, stillbirth, recurrent foetal growth retardation, or possible preeclampsia [24].
Despite national guidelines for thrombophilia testing, many women undergoing fertility treatment have frequently undergone thrombophilia testing as part of their investigation for infertility and some centres recommend that these tests should be a routine part of the work up prior to IVF [26]. This is due to the fact that adverse perinatal outcomes in women with identified thrombophilia have led to the suggestion that the inherited thrombophilia plays a role in subfertility, particularly recurrent implantation failure. However, due to the low prevalence of inherited thrombophilia, such studies have included small numbers of women and therefore, there is no clear answer as to whether inherited thrombophilia leads to reduced fertility [35]. Inherited thrombophilia has also been suggested to contribute to implantation failure following IVF, with an increased number of women with Factor V Leiden and prothrombin G20210A heterozygotes failing to conceive after three or more embryo transfers [36]. Some studies [37] have confirmed these findings but others have not [38]. Overall, it is possible that inherited thrombophilia plays a part in recurrent implantation failure, but this effect is likely to be small [35]. The role of low molecular weight heparins [LMWHs] in this population of women for improving fertility treatment outcome remains unclear.
The role of acquired thrombophilia in subfertility is also controversial. Antiphospholipid (aPL) antibodies are a heterogeneous group of antibodies directed against complexes of negatively charged phospholipids and their protein “cofactors.” They include lupus anticoagulant (LA), anticardiolipin (aCL) antibodies, antibodies against other phospholipids, as well as antibodies that react directly with the phospholipid-binding proteins (cofactors) such as β2glycoprotein I (β2GPI). With the exception of lupus anticoagulant, the other aPL antibodies are detected by immunoassays which have wide variability and poor standardization. The lupus anticoagulant is a coagulation-based assay which is more difficult to perform but is more reliable [39]. There has been found to be an increased incidence of antiphospholipid antibodies, with no evidence of antiphospholipid syndrome, in women with subfertility and those with recurrent IVF implantation failure (22 and 30 % respectively) than the healthy, fertile population (1–3 %). However, there are many pitfalls with these studies. The number of aPL epitopes evaluated varies between studies and many aPL antibodies that are not clinically relevant have been evaluated. There is no standardization of aPL assays, except for the aCL antibodies and LA, which can lead to a high degree of variability in the detection of aPL antibodies between laboratories. The cut-off point for a positive aPL antibody test also varies between studies. There is also a variation in the assays that have been used to detect LA [39]. Buckingham et al. conclude that the current weight of evidence suggests that aPL are more common in infertile women, especially those who have recurrent implantation failure than in control populations [39]. Despite this, the presence of aPL does not predict a poor outcome in IVF. However, this lack of predictive value may be due to the fact that many aPL that are not clinically relevant are being tested for, causing an overdiagnosis of aPL. The treatment of these women with antithrombotic therapy therefore remains empirical, and further studies using the widely accepted aPL antibodies (aCL antibodies, antiB2GPI, and LA) in infertile women would give a clearer picture of the role of these aPL antibodies in infertility.
It is clear that women diagnosed with antiphospholipid syndrome by laboratory and clinical criteria [40] do have poor pregnancy outcomes. Therefore, therapeutic or prophylactic low-molecular weight heparin (LMWH) and low-dose aspirin should be commenced at the time of ovarian stimulation and continued throughout pregnancy and for 6 weeks postpartum [1].
One randomized controlled trial provides further support to the view that prothrombotic changes may impact on implantation rates. This study reviewed 83 women with at least one thrombophilic defect (inherited or acquired), and three or more IVF failures. They were divided into two groups. The first group received prophylactic dose LMWH and the second group received placebo. The primary outcomes implantation, pregnancy, and live birth rate were significantly higher in the group receiving LMWH [41].
As well as the issue of subfertility, inherited and acquired thrombophilia also raises the question of need for thromboprophylaxis. The indications for thromboprophylaxis in the various thrombophilic defects are discussed further in this chapter in the section “Prevention of thrombosis.”
Prevention of Thrombosis
The Royal College of Obstetricians and Gynaecologists (RCOG) Green-top Guidelines suggest that all pregnant women should undergo a documented risk assessment prior to, or in, early pregnancy [42]. They suggest a scoring system dependent on the relative risk of various factors. The risk assessment tool in Table 4.2 has been adapted from the RCOG Green-top guideline No 37a.
Table 4.2
Risk assessment for thrombosis in pregnancy
Pre-existing risk factors | Tick if positive | Score |
|---|---|---|
Previous recurrent VTE | 3 | |
Previous VTE – unprovoked or oestrogen related | 3 | |
Previous VTE – provoked | 2 | |
Family history of VTE | 1 | |
Known thrombophilia | 2 | |
Medical comorbidities | 2 | |
Age (>35 years) | 1 | |
Obesity BMI >30 kg/m2 (based on booking weight) | 1 | |
BMI >40 kg/m2 (based on booking weight) | 2 | |
Parity > 3 | 1 | |
Smoker | 1 | |
Gross varicose veins | 1 | |
Obstetric risk factors | 1 | |
Pre-eclampsia | 1 | |
Dehydration/hyperemesis/OHSS | 1 | |
Multiple pregnancy or ART | 1
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|