REVIEW URRENT C OPINION
An update on the management of bleeding disorders during pregnancy Peter A. Kouides
Purpose of review The morbidity and mortality of postpartum hemorrhage (PPH) in women with an underlying bleeding disorder requires vigilance by the hematologist. Recent findings Recent studies suggest that women with underlying bleeding disorders may be ‘undertreated’ at the time of delivery in aiming for too low a target elevation that historically per numerous society guidelines has aimed for VWF/FVIII:C levels to be ‘only’ greater than 50% when in actuality the levels should be akin to what is achieved in a normal pregnancy. The result appears to be an increase in the rate and degree of PPH. In this context, although recent studies imply DDAVP is well tolerated, DDAVP may not be appropriate because it may not raise the levels into the normal supraphysiological range nor maintain it for several days. Particularly in women with rare bleeding disorders, i.e., non-FVIII:C or VWF deficient, adjunctive antifibrinolytic therapy, e.g., tranexamic acid, appears to be in order as a prophylactic measure. Summary Women with an underlying bleeding disorder appear to be at a heightened risk for PPH if the respective coagulation factor level is not appropriately replaced to the level that is physiologically achieved in a normal pregnancy. Furthermore, there appears to be underuse of tranexamic acid for prophylaxis of PPH in this population. Keywords bleeding disorders, postpartum hemorrhage, pregnancy, thrombocytopathy, von Willebrand disease
INTRODUCTION Women with an inherited bleeding disorder can have a relatively stable clinical course as there are numerous hemostatic and gynecological options in managing heavy menses with subsequent improvement in quality of life. The care of these patients can, however, be very challenging at the time of childbirth when the underlying coagulopathy precludes the normal supra-physiological compensatory rise that is protective for postpartum hemorrhage (PPH). This review will focus on recent epidemiological and treatment studies in pregnant women with an underlying inherited bleeding disorder.
The procoagulant, ‘protective’ state of pregnancy Near delivery, the VWF antigen (VWF:AG) and VWF ristocetin cofactor (VWF:RCo) activity peak in the 225–250% range [1]. Szecsi et al. [2] reported that the factor VIII activity at 38–42 weeks was 130–430% (2.5th and 97.5th percentiles) (n ¼ 73) compared with 82–291% (n-129) at 13–20-week gestation.
In the past 3 years, several studies have focused on the rate of fall of the FVIII:C and VWF levels postpartum. Huq et al. [3] carried out a prospective study assessing FVIII:C activity, VWF activity, and antigen levels in 95 women (with singleton uncomplicated pregnancies) during labor and postpartum on days 1, 2, and 3. They reported no significant differences in FVIII:C, VWF:Ag, and VWF:collagen binding activity on days 1 and 2 compared with levels during labor. There was a significant decrease in VWF:Ag (P ¼ 0.009) and VWF:collagen binding (P ¼ 0.04) on day 3 (Fig. 1a). Age, ethnicity, duration of labor, and mode of delivery did not have any significant effect on the changes in FVIII:C and VWF Mary M. Gooley Hemophilia Center, Inc. and University of Rochester School of Medicine, Rochester, New York, USA Correspondence to Peter A. Kouides, MD, Rochester General Hospital, 1425 Portland Avenue, Rochester, NY 14621, USA. Tel: +1 585 922 4020; fax: +1 585 563 1832; e-mail: [email protected] Curr Opin Hematol 2015, 22:397–405 DOI:10.1097/MOH.0000000000000167
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Hemostasis and thrombosis
KEY POINTS Although coagulant factors increase in the third trimester in women with underlying bleeding disorders, the rise is relatively lower than the supraphysiological elevation noted in normal pregnant women. Women with underlying bleeding disorders may be ‘undertreated’ at the time of delivery in aiming for too low a target elevation that historically per numerous society guidelines has aimed for VWF/FVIII:C levels to be ‘only’ greater than 50% when in actuality the levels should be akin to what is achieved in a normal pregnancy. The result appears to be an increase in the rate and degree of PPH. Recent studies imply DDAVP is well tolerated, but DDAVP may not be appropriate because it may not raise the levels into the normal supraphysiological range nor maintain it for several days. Particularly in women with rare bleeding disorders (RBD), i.e., non-FVIII:C or VWF deficient, adjunctive antifibrinolytic therapy, e.g., tranexamic acid, appears to be in order as a prophylactic measure although specific studies in this population are in order as currently we have to extrapolate from studies of tranexamic acid in normal pregnant women. The recent 2014 United Kingdom Haemophilia Centre Doctor’s Organization guideline for pregnancy management of rare bleeding disorders is a very useful document when encountering a pregnant RBD patient.
&
levels. Most recently, James et al. [4 ] not only studied comprehensively the rate of fall of FVIII:C and VWF levels postpartum in normal pregnant women but also in pregnant von Willebrand disease (VWD) patients. Patients were matched by race and age. In this prospective observational cohort study, VWF:RCo, VWF:Ag, and FVIII:C levels were obtained in the third trimester of pregnancy, on admission for childbirth, and 10 times postpartum over 6 weeks! Forty women (40 pregnancies) without VWD and 32 women (35 pregnancies) with well documented VWD were enrolled. A total of 15/32 with VWD were treated (30% of those with type 1 and all of those with type 2) in 17 pregnancies. The VWF levels peaked at 250% of baseline at 4 h postpartum in women with VWD and 12 h postpartum in women without VWD. Thereafter, VWF levels fell rapidly, approached baseline at 1 week, and reached baseline at 3 weeks (Fig. 1b). Interestingly, the FVIII:C level fell 20% at 24 h postpartum (perhaps because of consumption) and then rose 30% to a peak the next day (day 3 postpartum). Except immediately postpartum, when the levels among 398
www.co-hematology.com
treated patients were higher, levels among women with VWD appeared to parallel, but were lower than those among women without VWD. Interestingly, and to be discussed further in this review, levels were lowest among those who received treatment and this subgroup unlike the untreated type 1 VWD patients had a significantly increased estimated blood loss compared with the control group [4 ]. &
The hemorrhagic postpartum state of inherited bleeding disorders Furthermore, the clinician should be aware that not only may an underlying bleeding disorder be a cause of PPH but that there are additional causes that in general are more common, as outlined in Table 1. A recent novel approach using a candidate gene approach showed a significant association with PPH in a general population of women with PPH of the promoter polymorphism of the tissue factor gene (F3-603A > G) with the G allele exerting a protective effect (P ¼ 0.00053; OR ¼ 0.79, 95% confidence interval (CI) ¼ 0.69–0.90). The authors conclude that the protective effect against PPH of the TF-603A>G polymorphism is biologically plausible as the G allele is associated with an increased protein expression and tissue factor is strongly represented in the placenta at term, particularly in decidual cells of maternal origin [5 ]. This study is a reminder that PPH is ultimately a multifactorial process as outlined in Table 1, with future studies likely to reveal additional genetic polymorphisms that perhaps coupled with clinical risk factors and underlying inherited bleeding disorders result in major PPH. These risks notwithstanding, not surprisingly, given the inability of the inherited bleeding disorder patient to mount as high a supraphysiological rise in the coagulation levels peri-partum, historically a higher rate of PPH in women with primarily type 1 VWD of 16–29% [6–8] vs. 3–5% in the general population has been reported, in the first 24 h following delivery. Also, there appears to be a higher rate in type 2, 3 patients compared with type 1 patients [6,8,9]. These studies are from European and US centers and raise the question of whether similar outcomes are seen in non-Western countries. A recent retrospective single center study of 100 Iranian women with inherited bleeding disorders by Shahbazzi et al. [10] also reported an increased rate of PPH. Primary PPH was noted in 47.7% of hemophilia carriers, 32.1% of VWD patients, and 44.0% of patients with rare bleeding disorders compared with 7.5% in the control group of 200 women. Five percent of the patients required a red blood cell transfusion. Secondary PPH (defined as prolonged bleeding beyond 24 h postpartum up &
Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides
(a)
Factor level in IUdL
–1
370
320
270
220 Active labour
Day 1
Day 3
Time of sample Median VWF:Ag
Mean FVII:C
Mean VWF:CB
Median FVIII:C
Median VWF:CB
Median VWF:Ag
(b)
−10
(c)
300
VWF:RCo untreated type 1 200
VWF:RCo without VWD
150 100 50
0
10
20
30
Days postpartum
40
−10
VWF:Ag untreated type 1 VWF:Ag without VWD 200
100
FVIII IU dL–1
(b)
250
VWF:Ag IU dL–1
VWF:RCo IU dL–1
(a)
200
FVIII - untreated type 1 FVIII - without VWD
150
100
50
0
10
20
30
Days postpartum
40
−10
0
10
20
30
40
Days postpartum
FIGURE 1. (a) Mean and median factor levels in six normal patients who had consecutive samplings from Huq et al. [3]. (b) VWF:RCo, VWF Ag, and FVIII:C levels among women with untreated type 1 VWD compared with those among women without VWD from James et al. [4 ]. VWD, von Willebrand disease. &
to 6 weeks) was noted in 55.3% of hemophilia carriers, 28.6% of VWD patients, and 41.6% of patients with rare bleeding disorders compared with 12.6% in the control group [10]. These retrospective, single institution studies have, however, an inherent selection bias, whereas population-based studies have reported a lower rate of PPH [11]. Regarding the latter, one such study is an analysis of the United States Nationwide Inpatient Sample involving 4067 deliveries among women with VWD (1 in 4000 deliveries). James and Jamison [11] observed that women with VWD were more likely to experience a postpartum hemorrhage (OR, 1.5; 95% CI: 1.1–2.0), and had a fivefold increased risk of being transfused (OR, 4.7; 95% CI: 3.2–7.0). A more recent case–control study from Aberdeen, UK analyzed a total of 62 deliveries in 33 women with VWD compared with controls matched for age, year of delivery, and parity [12].
Primary PPH was reported in 12/62 (19.4%) deliveries in women with VWD and 16/124 (12.9%) controls. The unadjusted odds ratio (OR) for VWD as a risk factor for PPH was 1.62 (95% CI 0.75–3.49, P ¼ 0.22). After adjustment for other risk factors for PPH, the OR for VWD as a risk factor for PPH was 1.31 (95% CI 0.48–3.60, P ¼ 0.60). PPH was observed in 7/24 (29%) deliveries in women known prepregnancy to have VWD. The unadjusted odds for VWD as a risk factor for PPH in this group was significantly greater than the control group (OR 2.78 (95% CI 1.03–7.49), P ¼ 0.043) and remained significant after adjusting for other significant risk factors (OR 3.41 (95% CI 1.07–10.9), P ¼ 0.038). The authors concluded that VWD in itself may not be a significant risk factor for PPH; however, women known to have VWD predelivery may represent an at-risk subgroup consistent with the James and Jamison study [13].
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
399
Hemostasis and thrombosis Table 1. Multifactorial nature of Postpartum hemorrhage in terms of atony, trauma, placental abnormalities and hemostatic abnormalities Uterine atony Labor >2 h
Uterine/cervical/vaginal trauma Instrumental delivery C/S
Prolonged third stage
Hemostatic abnormalities
Multiple pregnancies
Acquired hemophilia
Induction of labor
Maternal obesity
Anticoagulant therapy
Baby >4 kg
Uremia, liver disease DIC: maternal sepsis, amniotic fluid embolism, placental abruption, massive transfusion
Placental abnormalities
Inherited
Retained placenta
VWD
Placenta previa
Thrombocytopathy
Placenta abruption VWD, von Willebrand disease.
MANAGEMENT OF INHERITED BLEEDING DISORDERS DURING PREGNANCY
in nonpregnant patients, hyponatremia 130 meq/l or less was noted at a relatively high rate of 11% (11/ 107; 95% CI 5–16%) after a preoperative dose [16 ]. Consequently, caution must be exercised when administering DDAVP peri-partum and to be discussed below factor replacement may be preferable in VWD and hemophilia A in achieving a higher physiological target level noted in normal pregnant patients postpartum. &
For an overview, see Fig. 2.
von Willebrand disease and hemophilia carriers The use of desmopressin In cases in which the VWF and or FVIII:C levels have not normalized by the time of active labor, there may be a tight correlation between certain genotypes and biological responsiveness to DDAVP and the risk of postpartum bleeding [14]. Historically, for women whose levels have not exceeded 50%, there are, however, theoretical concerns for administering DDAVP prepartum for the following reasons: (1) vaso-constrictive effect leading to decreased placental flow, (2) risk of premature labor as desmopressin has weak V2 receptor activity, (3) risk of neonatal hyponatremia. A systematic review of 30 studies of the use of DDAVP for treatment and prophylaxis of bleeding disorders in pregnancy, however, further confirmed its efficacy and safety [15]. They did collect two cases of symptomatic hyponatremia postpartum in 172 pregnancies. Only 5 of 172 patients (3%) of peripartum use were associated with adverse bleeding in two patients with Hermansky Pudlak Syndrome, one case of Ehler Danos, one case of storage pool deficiency, and one case of VWD. This review also included 51 patients in the first or second trimester with no reports of toxicity or adverse bleeding [15]. Most reports have not systematically recorded postpartum sodium levels after administration. Recently 400
www.co-hematology.com
Replacement therapy in von Willebrand disease or in hemophilia carriers Numerous societies and national organizations have developed guidelines as outlined in Table 2 [17–19,20 ]. They all consistently state that in the third trimester, the respective level (VWF, FVIII:C) must be greater than 50% to both permit safely epidural analgesia but also to prevent PPH. In those patients who are not desmopressin (DDAVP) responsive, however, which is essentially most patients who by the third trimester do not have a factor VIII or VWF level greater than 50%, specific guidance is lacking in terms of how high a level should be achieved and for how long. Specifically, guidance is lacking on whether a level in the 150–200% range should be targeted with replacement therapy; a range in which levels typically are achieved during a normal pregnancy [1,2,21] as reviewed earlier in this monograph. Furthermore, as noted, VWF levels both in normal patients [3] and VWD patients [4 ] can fall rapidly approaching baseline in 1 week and reach baseline at 3 weeks. In addition, the various present guidelines by and large do not state how long replacement therapy should be given. Some groups also state that delivery should be managed at a tertiary center [18,20 ]. None of the groups specifically advises prophylactic antifibrinolytic therapy &
&
&
Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides
Clinical peri-partum scenario for the bleeding disorder patient
Prophylaxis of PPH
If VWD or FVIII carrier
Treatment of PPH
If Rare bleeding disorder
If Thrombocytopathy
rd
If 3 trimester level < 50%, replace to respective level >100–200% peripartum then minimum 4–7 days post-partum (if prior robust responder to DDAVP could consider that instead of replacement)
Consider peripartum IV TA and post discharge oral TA or EACA if high bleeding score
Specific factor replacement per UK 2014 guidelines in Table 3
Peripartum IV TA and post discharge oral TA or EACA
IV DDAVP + IV TA and postdischarge oral TA or EACA
Platelet transfusion at C/S or if prior history of major bleeding at time of active labor
1. Fundal massage 2. Crystalloid resuscitation and red cells as indicated 3. Double utertonicse.g. oxytocin and misoprostol 4. Continue factor replacement and IV TA 5. Continued obstetrical assessment and possible intervention 6. Fibrinogen replacement for superimposed DIC 7. rVIIa before hysterectomy
FIGURE 2. General management approach for prophylaxis and treatment of PPH in a patient with an underlying bleeding disorder. PPH, postpartum hemorrhage; VWD, von Willebrand disease.
postpartum even though bleeding can occur up to 4–6 weeks. The national guidelines in The Netherlands have not been that different from other societies advising a target level of 100% level and then a follow-up dose of half the first dose 24 h later. They also advise consultation at a center of specialized care, in other words a Hemophilia Treatment Center, but they do not advise specifically that the delivery be carried out at that specialized center but rather the advisements from that consultation be carried out [22 ]. Published his year, Stoof et al. [22 ], on behalf of three hemophilia treatment centers in The Netherlands, assessed by retrospective study whether such advisement indeed translates to a normal delivery without excessive PPH. The authors reviewed records of 185 deliveries in 154 women with either previously documented VWD or hemophilia carriership that delivered over a 9-year period. The main endpoint was primary PPH that they defined 500 ml within 24 h postpartum and severe PPH as blood loss at least 1000 ml. Approximately a third of the deliveries were associated with primary PPH and approximately a third of those patients had severe PPH. The authors noted an &&
&&
inverse relationship between the incidence of PPH and third trimester factor levels. These results are quite provocative, consistent with the results of James et al. [4 ] already mentioned. In that study, the mean estimated blood loss (EBL) at delivery for treated women was 615 ml (473, 758), which was significantly greater than the mean for other women whose mean EBL was 448 ml (379, 517) (P < 0.05). In essence, these women were undertreated [4 ]. In that study, all the patients had VWD, but in the Stoof et al. study, some of these women were hemophilia carriers. Interestingly, PPH was noted in that cohort also in 35 of 114 (31%) pregnancies. In the 14 pregnancies with severe PPH, half had VWD and the remaining half were carriers; five of seven of the carriers had hemophilia A. In the hemophilia B carriers, the third trimester factor IX level, respectively, was 17% and 75% and for the hemophilia A carriers, third trimester factor VIII levels were available on four of five and were 112–165% with the respective nonpregnant factor VIII levels being 50–136%. It is fascinating that there was still severe PPH in these carriers despite the levels being at least 50%. It is possible that these patients are not able to ‘mount’ a
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
&
&
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
401
Hemostasis and thrombosis Table 2. Guidelines for preventing postpartum hemorrhage in von Willebrand disease Organization/year
Advisement
Canadian 2005 [17]
‘It is generally recommended to keep factor levels above 50% for 3–4 days after vaginal delivery and 4–5 days after a cesarean section. Women at risk of late postpartum hemorrhage should have their hemoglobin checked before discharge. They should be instructed about possible excessive bleeding and should be seen in a follow-up visit 2-weeks postpartum’.
NHLBI 2008 [18]
‘Should be referred to a center that has high-risk obstetrics capabilities and expertise in haemostasis for prenatal care, delivery, termination of pregnancy, or management of miscarriage. Grade C, level IV. Should achieve VWF:RCo and FVIII levels at least 50% before delivery and maintain those levels for at least 3–5 days afterward, Grade C, level IV’.
Italian 2009 [19]
‘The risk of bleeding is minimal when FVIII:C and VWF:RCo levels are higher than 30 IU/dL. In order to prevent late bleeding, VWF:RCo and FVIII:C levels should be checked and women monitored clinically for at least 2 weeks postpartum’.
UK 2014 [20]
‘Women with VWD whose VWF activity does not rise to normal levels during pregnancy should be delivered in an obstetric unit that can easily and quickly access Haemophilia. Center and comprehensive neonatal care facilities. Von Willebrand factor parameters should be checked at booking and at around 34-week gestation unless levels have already been shown to have risen to normal. Neuraxial anaesthesia, vaginal delivery, and Caesarean section can all be regarded as well tolerated in type 1 when VWF:RCo > 0.50 iu/ml. Patients may require several days’ treatment, especially after Caesarean section. Even if VWF parameters are satisfactory at the time of delivery, abnormal bleeding may ensue following discharge from hospital. It is important that women are made aware of this and advised to seek appropriate medical help should they develop persistent heavy postpartum bleeding’.
VWD, von Willebrand disease.
level that in normal patients would be protective for hemorrhage. Indirect evidence includes studies in carriers with prolonged bleeding and morbidity with procedures [23,24 ] as well as decreased range of motion and magnetic resonance imaging evidence of joint damage [25] and decreased quality of life despite levels above 40% [26]. Stoof et al.’s study is noteworthy in questioning the ‘status quo’ that specialized care is effective, that a postpartum postreplacement factor VIII level of 100% is adequate, that administration of half the initial dose as follow-up treatment is adequate, and that the obstetric care of hemophilia carriers does not require consideration for aggressive replacement therapy. These results are definitely hypothesis generating, but the clinical researcher in this field must consider several limitations of the Stoof et al. study – the study design was retrospective, there was seemingly a high rate of induction of labor of 34% (though it should be pointed out that induction of labor is not a standard procedure in women with bleeding disorders nor is advised in the Dutch national guidelines), and there were 42 patients without information. Nonetheless, their study is provocative given the high rate of PPH despite treatment. These results should compel one to consider using concurrent antifibrinolytic agent therapy [27 ] and/or double utertonics (oxytocin þ misoprostol) [28 ] and/or aiming for a higher trough factor level closer to 200% then 100% and maintaining such a level for several days postpartum (with the implications, for &
&
&
402
www.co-hematology.com
better or worse, being the need to transfer women to specialized centers capable to do on site factor levels). One could also argue that desmopressin should not be used at the time of active labor even apart from the practical concern of inducing hyponatremia as it may not dependably raise the factor level in the physiologic range nor of course maintain it for several days. Recently, a case of severe hemophilia (0.9%) managed successfully in two pregnancies was reported beginning prophylaxis with recombinant FVIII:C 2 week 40 u/kg with trough of 7% followed by immediate postpartum infusions to target FVIII:C more than 100% followed by prophylaxis 6 weeks postpartum initially 25 U/kg every 12 h days 2–5 then daily days 6–14 then every other day until week 6 [29].
Antifibrinolytic therapy in women with underlying bleeding disorders As mentioned in the preceding section, antifibrinolytic therapy can be an option for the prevention and control of PPH on the basis of the observation that at the time of delivery, the synthesis of plasminogen activator inhibitor-2 ceases with concurrent elevation of tissue plasminogen activator leading to subsequent fibrin and fibrinogen degradation [30]. Unfortunately, there have not been any randomized control trials of tranexamic acid for the prevention or treatment of tranexamic acid in women with underlying inherited bleeding disorders [31]. One Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides Table 3. 2014 United Kingdom Haemophilia Centre Doctor’s Organization guidelines for prophylaxis for postpartum hemorrhage in women with an inherited bleeding disorder [34 ] &&
Deficiency state
Recommendation
Fibrinogen
For women with fibrinogen activity 1 g/l. Consider additional fibrinogen concentrate at established labor to ensure fibrinogen activity >1_5 g/l for at least 3 days (2C). For pregnant women with thrombotic dysfibrinogenaemia or with afibrinogenaemia or hypofibrinogenaemia and other risk factors for venous thrombosis, consider thromboprophylaxis with low molecular weight heparin (2C).
Factor II
For delivery in women with FII activity 0.2 iu/ml for at least 3 days (2C). SD-FFP 15–25 ml/kg is an alternative if PCC is unavailable (2C).
Factor V
For delivery in women with FV activity < 0.2 iu/ml, consider SD-FFP 15–25 ml/kg once in established labour or before caesarean section, to achieve FV activity 0.2–0.4 iu/ml. Consider further SD-FFP 10 ml/kg at 12-h intervals to maintain FV activity >0_2 iu/ml for at least 3 days (2C).
Factor VII
For delivery in women with FVII activity 0_3 iu/ml for at least 3 days (2C).
Factor XI
For delivery in all women with factor XI activity
An update on the management of bleeding disorders during pregnancy Peter A. Kouides
Purpose of review The morbidity and mortality of postpartum hemorrhage (PPH) in women with an underlying bleeding disorder requires vigilance by the hematologist. Recent findings Recent studies suggest that women with underlying bleeding disorders may be ‘undertreated’ at the time of delivery in aiming for too low a target elevation that historically per numerous society guidelines has aimed for VWF/FVIII:C levels to be ‘only’ greater than 50% when in actuality the levels should be akin to what is achieved in a normal pregnancy. The result appears to be an increase in the rate and degree of PPH. In this context, although recent studies imply DDAVP is well tolerated, DDAVP may not be appropriate because it may not raise the levels into the normal supraphysiological range nor maintain it for several days. Particularly in women with rare bleeding disorders, i.e., non-FVIII:C or VWF deficient, adjunctive antifibrinolytic therapy, e.g., tranexamic acid, appears to be in order as a prophylactic measure. Summary Women with an underlying bleeding disorder appear to be at a heightened risk for PPH if the respective coagulation factor level is not appropriately replaced to the level that is physiologically achieved in a normal pregnancy. Furthermore, there appears to be underuse of tranexamic acid for prophylaxis of PPH in this population. Keywords bleeding disorders, postpartum hemorrhage, pregnancy, thrombocytopathy, von Willebrand disease
INTRODUCTION Women with an inherited bleeding disorder can have a relatively stable clinical course as there are numerous hemostatic and gynecological options in managing heavy menses with subsequent improvement in quality of life. The care of these patients can, however, be very challenging at the time of childbirth when the underlying coagulopathy precludes the normal supra-physiological compensatory rise that is protective for postpartum hemorrhage (PPH). This review will focus on recent epidemiological and treatment studies in pregnant women with an underlying inherited bleeding disorder.
The procoagulant, ‘protective’ state of pregnancy Near delivery, the VWF antigen (VWF:AG) and VWF ristocetin cofactor (VWF:RCo) activity peak in the 225–250% range [1]. Szecsi et al. [2] reported that the factor VIII activity at 38–42 weeks was 130–430% (2.5th and 97.5th percentiles) (n ¼ 73) compared with 82–291% (n-129) at 13–20-week gestation.
In the past 3 years, several studies have focused on the rate of fall of the FVIII:C and VWF levels postpartum. Huq et al. [3] carried out a prospective study assessing FVIII:C activity, VWF activity, and antigen levels in 95 women (with singleton uncomplicated pregnancies) during labor and postpartum on days 1, 2, and 3. They reported no significant differences in FVIII:C, VWF:Ag, and VWF:collagen binding activity on days 1 and 2 compared with levels during labor. There was a significant decrease in VWF:Ag (P ¼ 0.009) and VWF:collagen binding (P ¼ 0.04) on day 3 (Fig. 1a). Age, ethnicity, duration of labor, and mode of delivery did not have any significant effect on the changes in FVIII:C and VWF Mary M. Gooley Hemophilia Center, Inc. and University of Rochester School of Medicine, Rochester, New York, USA Correspondence to Peter A. Kouides, MD, Rochester General Hospital, 1425 Portland Avenue, Rochester, NY 14621, USA. Tel: +1 585 922 4020; fax: +1 585 563 1832; e-mail: [email protected] Curr Opin Hematol 2015, 22:397–405 DOI:10.1097/MOH.0000000000000167
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Hemostasis and thrombosis
KEY POINTS Although coagulant factors increase in the third trimester in women with underlying bleeding disorders, the rise is relatively lower than the supraphysiological elevation noted in normal pregnant women. Women with underlying bleeding disorders may be ‘undertreated’ at the time of delivery in aiming for too low a target elevation that historically per numerous society guidelines has aimed for VWF/FVIII:C levels to be ‘only’ greater than 50% when in actuality the levels should be akin to what is achieved in a normal pregnancy. The result appears to be an increase in the rate and degree of PPH. Recent studies imply DDAVP is well tolerated, but DDAVP may not be appropriate because it may not raise the levels into the normal supraphysiological range nor maintain it for several days. Particularly in women with rare bleeding disorders (RBD), i.e., non-FVIII:C or VWF deficient, adjunctive antifibrinolytic therapy, e.g., tranexamic acid, appears to be in order as a prophylactic measure although specific studies in this population are in order as currently we have to extrapolate from studies of tranexamic acid in normal pregnant women. The recent 2014 United Kingdom Haemophilia Centre Doctor’s Organization guideline for pregnancy management of rare bleeding disorders is a very useful document when encountering a pregnant RBD patient.
&
levels. Most recently, James et al. [4 ] not only studied comprehensively the rate of fall of FVIII:C and VWF levels postpartum in normal pregnant women but also in pregnant von Willebrand disease (VWD) patients. Patients were matched by race and age. In this prospective observational cohort study, VWF:RCo, VWF:Ag, and FVIII:C levels were obtained in the third trimester of pregnancy, on admission for childbirth, and 10 times postpartum over 6 weeks! Forty women (40 pregnancies) without VWD and 32 women (35 pregnancies) with well documented VWD were enrolled. A total of 15/32 with VWD were treated (30% of those with type 1 and all of those with type 2) in 17 pregnancies. The VWF levels peaked at 250% of baseline at 4 h postpartum in women with VWD and 12 h postpartum in women without VWD. Thereafter, VWF levels fell rapidly, approached baseline at 1 week, and reached baseline at 3 weeks (Fig. 1b). Interestingly, the FVIII:C level fell 20% at 24 h postpartum (perhaps because of consumption) and then rose 30% to a peak the next day (day 3 postpartum). Except immediately postpartum, when the levels among 398
www.co-hematology.com
treated patients were higher, levels among women with VWD appeared to parallel, but were lower than those among women without VWD. Interestingly, and to be discussed further in this review, levels were lowest among those who received treatment and this subgroup unlike the untreated type 1 VWD patients had a significantly increased estimated blood loss compared with the control group [4 ]. &
The hemorrhagic postpartum state of inherited bleeding disorders Furthermore, the clinician should be aware that not only may an underlying bleeding disorder be a cause of PPH but that there are additional causes that in general are more common, as outlined in Table 1. A recent novel approach using a candidate gene approach showed a significant association with PPH in a general population of women with PPH of the promoter polymorphism of the tissue factor gene (F3-603A > G) with the G allele exerting a protective effect (P ¼ 0.00053; OR ¼ 0.79, 95% confidence interval (CI) ¼ 0.69–0.90). The authors conclude that the protective effect against PPH of the TF-603A>G polymorphism is biologically plausible as the G allele is associated with an increased protein expression and tissue factor is strongly represented in the placenta at term, particularly in decidual cells of maternal origin [5 ]. This study is a reminder that PPH is ultimately a multifactorial process as outlined in Table 1, with future studies likely to reveal additional genetic polymorphisms that perhaps coupled with clinical risk factors and underlying inherited bleeding disorders result in major PPH. These risks notwithstanding, not surprisingly, given the inability of the inherited bleeding disorder patient to mount as high a supraphysiological rise in the coagulation levels peri-partum, historically a higher rate of PPH in women with primarily type 1 VWD of 16–29% [6–8] vs. 3–5% in the general population has been reported, in the first 24 h following delivery. Also, there appears to be a higher rate in type 2, 3 patients compared with type 1 patients [6,8,9]. These studies are from European and US centers and raise the question of whether similar outcomes are seen in non-Western countries. A recent retrospective single center study of 100 Iranian women with inherited bleeding disorders by Shahbazzi et al. [10] also reported an increased rate of PPH. Primary PPH was noted in 47.7% of hemophilia carriers, 32.1% of VWD patients, and 44.0% of patients with rare bleeding disorders compared with 7.5% in the control group of 200 women. Five percent of the patients required a red blood cell transfusion. Secondary PPH (defined as prolonged bleeding beyond 24 h postpartum up &
Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides
(a)
Factor level in IUdL
–1
370
320
270
220 Active labour
Day 1
Day 3
Time of sample Median VWF:Ag
Mean FVII:C
Mean VWF:CB
Median FVIII:C
Median VWF:CB
Median VWF:Ag
(b)
−10
(c)
300
VWF:RCo untreated type 1 200
VWF:RCo without VWD
150 100 50
0
10
20
30
Days postpartum
40
−10
VWF:Ag untreated type 1 VWF:Ag without VWD 200
100
FVIII IU dL–1
(b)
250
VWF:Ag IU dL–1
VWF:RCo IU dL–1
(a)
200
FVIII - untreated type 1 FVIII - without VWD
150
100
50
0
10
20
30
Days postpartum
40
−10
0
10
20
30
40
Days postpartum
FIGURE 1. (a) Mean and median factor levels in six normal patients who had consecutive samplings from Huq et al. [3]. (b) VWF:RCo, VWF Ag, and FVIII:C levels among women with untreated type 1 VWD compared with those among women without VWD from James et al. [4 ]. VWD, von Willebrand disease. &
to 6 weeks) was noted in 55.3% of hemophilia carriers, 28.6% of VWD patients, and 41.6% of patients with rare bleeding disorders compared with 12.6% in the control group [10]. These retrospective, single institution studies have, however, an inherent selection bias, whereas population-based studies have reported a lower rate of PPH [11]. Regarding the latter, one such study is an analysis of the United States Nationwide Inpatient Sample involving 4067 deliveries among women with VWD (1 in 4000 deliveries). James and Jamison [11] observed that women with VWD were more likely to experience a postpartum hemorrhage (OR, 1.5; 95% CI: 1.1–2.0), and had a fivefold increased risk of being transfused (OR, 4.7; 95% CI: 3.2–7.0). A more recent case–control study from Aberdeen, UK analyzed a total of 62 deliveries in 33 women with VWD compared with controls matched for age, year of delivery, and parity [12].
Primary PPH was reported in 12/62 (19.4%) deliveries in women with VWD and 16/124 (12.9%) controls. The unadjusted odds ratio (OR) for VWD as a risk factor for PPH was 1.62 (95% CI 0.75–3.49, P ¼ 0.22). After adjustment for other risk factors for PPH, the OR for VWD as a risk factor for PPH was 1.31 (95% CI 0.48–3.60, P ¼ 0.60). PPH was observed in 7/24 (29%) deliveries in women known prepregnancy to have VWD. The unadjusted odds for VWD as a risk factor for PPH in this group was significantly greater than the control group (OR 2.78 (95% CI 1.03–7.49), P ¼ 0.043) and remained significant after adjusting for other significant risk factors (OR 3.41 (95% CI 1.07–10.9), P ¼ 0.038). The authors concluded that VWD in itself may not be a significant risk factor for PPH; however, women known to have VWD predelivery may represent an at-risk subgroup consistent with the James and Jamison study [13].
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
399
Hemostasis and thrombosis Table 1. Multifactorial nature of Postpartum hemorrhage in terms of atony, trauma, placental abnormalities and hemostatic abnormalities Uterine atony Labor >2 h
Uterine/cervical/vaginal trauma Instrumental delivery C/S
Prolonged third stage
Hemostatic abnormalities
Multiple pregnancies
Acquired hemophilia
Induction of labor
Maternal obesity
Anticoagulant therapy
Baby >4 kg
Uremia, liver disease DIC: maternal sepsis, amniotic fluid embolism, placental abruption, massive transfusion
Placental abnormalities
Inherited
Retained placenta
VWD
Placenta previa
Thrombocytopathy
Placenta abruption VWD, von Willebrand disease.
MANAGEMENT OF INHERITED BLEEDING DISORDERS DURING PREGNANCY
in nonpregnant patients, hyponatremia 130 meq/l or less was noted at a relatively high rate of 11% (11/ 107; 95% CI 5–16%) after a preoperative dose [16 ]. Consequently, caution must be exercised when administering DDAVP peri-partum and to be discussed below factor replacement may be preferable in VWD and hemophilia A in achieving a higher physiological target level noted in normal pregnant patients postpartum. &
For an overview, see Fig. 2.
von Willebrand disease and hemophilia carriers The use of desmopressin In cases in which the VWF and or FVIII:C levels have not normalized by the time of active labor, there may be a tight correlation between certain genotypes and biological responsiveness to DDAVP and the risk of postpartum bleeding [14]. Historically, for women whose levels have not exceeded 50%, there are, however, theoretical concerns for administering DDAVP prepartum for the following reasons: (1) vaso-constrictive effect leading to decreased placental flow, (2) risk of premature labor as desmopressin has weak V2 receptor activity, (3) risk of neonatal hyponatremia. A systematic review of 30 studies of the use of DDAVP for treatment and prophylaxis of bleeding disorders in pregnancy, however, further confirmed its efficacy and safety [15]. They did collect two cases of symptomatic hyponatremia postpartum in 172 pregnancies. Only 5 of 172 patients (3%) of peripartum use were associated with adverse bleeding in two patients with Hermansky Pudlak Syndrome, one case of Ehler Danos, one case of storage pool deficiency, and one case of VWD. This review also included 51 patients in the first or second trimester with no reports of toxicity or adverse bleeding [15]. Most reports have not systematically recorded postpartum sodium levels after administration. Recently 400
www.co-hematology.com
Replacement therapy in von Willebrand disease or in hemophilia carriers Numerous societies and national organizations have developed guidelines as outlined in Table 2 [17–19,20 ]. They all consistently state that in the third trimester, the respective level (VWF, FVIII:C) must be greater than 50% to both permit safely epidural analgesia but also to prevent PPH. In those patients who are not desmopressin (DDAVP) responsive, however, which is essentially most patients who by the third trimester do not have a factor VIII or VWF level greater than 50%, specific guidance is lacking in terms of how high a level should be achieved and for how long. Specifically, guidance is lacking on whether a level in the 150–200% range should be targeted with replacement therapy; a range in which levels typically are achieved during a normal pregnancy [1,2,21] as reviewed earlier in this monograph. Furthermore, as noted, VWF levels both in normal patients [3] and VWD patients [4 ] can fall rapidly approaching baseline in 1 week and reach baseline at 3 weeks. In addition, the various present guidelines by and large do not state how long replacement therapy should be given. Some groups also state that delivery should be managed at a tertiary center [18,20 ]. None of the groups specifically advises prophylactic antifibrinolytic therapy &
&
&
Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides
Clinical peri-partum scenario for the bleeding disorder patient
Prophylaxis of PPH
If VWD or FVIII carrier
Treatment of PPH
If Rare bleeding disorder
If Thrombocytopathy
rd
If 3 trimester level < 50%, replace to respective level >100–200% peripartum then minimum 4–7 days post-partum (if prior robust responder to DDAVP could consider that instead of replacement)
Consider peripartum IV TA and post discharge oral TA or EACA if high bleeding score
Specific factor replacement per UK 2014 guidelines in Table 3
Peripartum IV TA and post discharge oral TA or EACA
IV DDAVP + IV TA and postdischarge oral TA or EACA
Platelet transfusion at C/S or if prior history of major bleeding at time of active labor
1. Fundal massage 2. Crystalloid resuscitation and red cells as indicated 3. Double utertonicse.g. oxytocin and misoprostol 4. Continue factor replacement and IV TA 5. Continued obstetrical assessment and possible intervention 6. Fibrinogen replacement for superimposed DIC 7. rVIIa before hysterectomy
FIGURE 2. General management approach for prophylaxis and treatment of PPH in a patient with an underlying bleeding disorder. PPH, postpartum hemorrhage; VWD, von Willebrand disease.
postpartum even though bleeding can occur up to 4–6 weeks. The national guidelines in The Netherlands have not been that different from other societies advising a target level of 100% level and then a follow-up dose of half the first dose 24 h later. They also advise consultation at a center of specialized care, in other words a Hemophilia Treatment Center, but they do not advise specifically that the delivery be carried out at that specialized center but rather the advisements from that consultation be carried out [22 ]. Published his year, Stoof et al. [22 ], on behalf of three hemophilia treatment centers in The Netherlands, assessed by retrospective study whether such advisement indeed translates to a normal delivery without excessive PPH. The authors reviewed records of 185 deliveries in 154 women with either previously documented VWD or hemophilia carriership that delivered over a 9-year period. The main endpoint was primary PPH that they defined 500 ml within 24 h postpartum and severe PPH as blood loss at least 1000 ml. Approximately a third of the deliveries were associated with primary PPH and approximately a third of those patients had severe PPH. The authors noted an &&
&&
inverse relationship between the incidence of PPH and third trimester factor levels. These results are quite provocative, consistent with the results of James et al. [4 ] already mentioned. In that study, the mean estimated blood loss (EBL) at delivery for treated women was 615 ml (473, 758), which was significantly greater than the mean for other women whose mean EBL was 448 ml (379, 517) (P < 0.05). In essence, these women were undertreated [4 ]. In that study, all the patients had VWD, but in the Stoof et al. study, some of these women were hemophilia carriers. Interestingly, PPH was noted in that cohort also in 35 of 114 (31%) pregnancies. In the 14 pregnancies with severe PPH, half had VWD and the remaining half were carriers; five of seven of the carriers had hemophilia A. In the hemophilia B carriers, the third trimester factor IX level, respectively, was 17% and 75% and for the hemophilia A carriers, third trimester factor VIII levels were available on four of five and were 112–165% with the respective nonpregnant factor VIII levels being 50–136%. It is fascinating that there was still severe PPH in these carriers despite the levels being at least 50%. It is possible that these patients are not able to ‘mount’ a
1065-6251 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved.
&
&
www.co-hematology.com
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
401
Hemostasis and thrombosis Table 2. Guidelines for preventing postpartum hemorrhage in von Willebrand disease Organization/year
Advisement
Canadian 2005 [17]
‘It is generally recommended to keep factor levels above 50% for 3–4 days after vaginal delivery and 4–5 days after a cesarean section. Women at risk of late postpartum hemorrhage should have their hemoglobin checked before discharge. They should be instructed about possible excessive bleeding and should be seen in a follow-up visit 2-weeks postpartum’.
NHLBI 2008 [18]
‘Should be referred to a center that has high-risk obstetrics capabilities and expertise in haemostasis for prenatal care, delivery, termination of pregnancy, or management of miscarriage. Grade C, level IV. Should achieve VWF:RCo and FVIII levels at least 50% before delivery and maintain those levels for at least 3–5 days afterward, Grade C, level IV’.
Italian 2009 [19]
‘The risk of bleeding is minimal when FVIII:C and VWF:RCo levels are higher than 30 IU/dL. In order to prevent late bleeding, VWF:RCo and FVIII:C levels should be checked and women monitored clinically for at least 2 weeks postpartum’.
UK 2014 [20]
‘Women with VWD whose VWF activity does not rise to normal levels during pregnancy should be delivered in an obstetric unit that can easily and quickly access Haemophilia. Center and comprehensive neonatal care facilities. Von Willebrand factor parameters should be checked at booking and at around 34-week gestation unless levels have already been shown to have risen to normal. Neuraxial anaesthesia, vaginal delivery, and Caesarean section can all be regarded as well tolerated in type 1 when VWF:RCo > 0.50 iu/ml. Patients may require several days’ treatment, especially after Caesarean section. Even if VWF parameters are satisfactory at the time of delivery, abnormal bleeding may ensue following discharge from hospital. It is important that women are made aware of this and advised to seek appropriate medical help should they develop persistent heavy postpartum bleeding’.
VWD, von Willebrand disease.
level that in normal patients would be protective for hemorrhage. Indirect evidence includes studies in carriers with prolonged bleeding and morbidity with procedures [23,24 ] as well as decreased range of motion and magnetic resonance imaging evidence of joint damage [25] and decreased quality of life despite levels above 40% [26]. Stoof et al.’s study is noteworthy in questioning the ‘status quo’ that specialized care is effective, that a postpartum postreplacement factor VIII level of 100% is adequate, that administration of half the initial dose as follow-up treatment is adequate, and that the obstetric care of hemophilia carriers does not require consideration for aggressive replacement therapy. These results are definitely hypothesis generating, but the clinical researcher in this field must consider several limitations of the Stoof et al. study – the study design was retrospective, there was seemingly a high rate of induction of labor of 34% (though it should be pointed out that induction of labor is not a standard procedure in women with bleeding disorders nor is advised in the Dutch national guidelines), and there were 42 patients without information. Nonetheless, their study is provocative given the high rate of PPH despite treatment. These results should compel one to consider using concurrent antifibrinolytic agent therapy [27 ] and/or double utertonics (oxytocin þ misoprostol) [28 ] and/or aiming for a higher trough factor level closer to 200% then 100% and maintaining such a level for several days postpartum (with the implications, for &
&
&
402
www.co-hematology.com
better or worse, being the need to transfer women to specialized centers capable to do on site factor levels). One could also argue that desmopressin should not be used at the time of active labor even apart from the practical concern of inducing hyponatremia as it may not dependably raise the factor level in the physiologic range nor of course maintain it for several days. Recently, a case of severe hemophilia (0.9%) managed successfully in two pregnancies was reported beginning prophylaxis with recombinant FVIII:C 2 week 40 u/kg with trough of 7% followed by immediate postpartum infusions to target FVIII:C more than 100% followed by prophylaxis 6 weeks postpartum initially 25 U/kg every 12 h days 2–5 then daily days 6–14 then every other day until week 6 [29].
Antifibrinolytic therapy in women with underlying bleeding disorders As mentioned in the preceding section, antifibrinolytic therapy can be an option for the prevention and control of PPH on the basis of the observation that at the time of delivery, the synthesis of plasminogen activator inhibitor-2 ceases with concurrent elevation of tissue plasminogen activator leading to subsequent fibrin and fibrinogen degradation [30]. Unfortunately, there have not been any randomized control trials of tranexamic acid for the prevention or treatment of tranexamic acid in women with underlying inherited bleeding disorders [31]. One Volume 22 Number 5 September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Management of bleeding disorders during pregnancy Kouides Table 3. 2014 United Kingdom Haemophilia Centre Doctor’s Organization guidelines for prophylaxis for postpartum hemorrhage in women with an inherited bleeding disorder [34 ] &&
Deficiency state
Recommendation
Fibrinogen
For women with fibrinogen activity 1 g/l. Consider additional fibrinogen concentrate at established labor to ensure fibrinogen activity >1_5 g/l for at least 3 days (2C). For pregnant women with thrombotic dysfibrinogenaemia or with afibrinogenaemia or hypofibrinogenaemia and other risk factors for venous thrombosis, consider thromboprophylaxis with low molecular weight heparin (2C).
Factor II
For delivery in women with FII activity 0.2 iu/ml for at least 3 days (2C). SD-FFP 15–25 ml/kg is an alternative if PCC is unavailable (2C).
Factor V
For delivery in women with FV activity < 0.2 iu/ml, consider SD-FFP 15–25 ml/kg once in established labour or before caesarean section, to achieve FV activity 0.2–0.4 iu/ml. Consider further SD-FFP 10 ml/kg at 12-h intervals to maintain FV activity >0_2 iu/ml for at least 3 days (2C).
Factor VII
For delivery in women with FVII activity 0_3 iu/ml for at least 3 days (2C).
Factor XI
For delivery in all women with factor XI activity
