Women and bleeding disorders

Authors


Andra H. James, MD, Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Duke University Medical Center, Box 3967 DUMC, Durham, NC 27710, USA.
Tel.: +1 919 668 0011; fax: +1 919 681 7861;
e-mail: andra.james@duke.edu

Abstract

Summary.  While women are rarely affected by haemophilia, they are equally as likely as men to have other bleeding disorders. Menorrhagia, or heavy menstrual bleeding, is the most common symptom that they experience. Not only is menorrhagia more prevalent among women with bleeding disorders, but bleeding disorders are more prevalent among women with menorrhagia. Although menorrhagia is the most common reproductive tract manifestation of a bleeding disorder, it is not the only manifestation. Women with bleeding disorders appear to be at an increased risk of developing haemorrhagic ovarian cysts and possibly endometriosis. Women suspected of having a bleeding disorder or being a carrier of haemophilia should be offered diagnostic testing before getting pregnant to allow for appropriate preconception counselling and pregnancy management. During pregnancy, women with bleeding disorders may be at an increased risk of bleeding complications. At the time of childbirth, women with bleeding disorders appear to be more likely to experience postpartum haemorrhage, particularly delayed or secondary postpartum haemorrhage. As women with bleeding disorders grow older, they may be more likely to manifest gynaecological conditions which present with bleeding. Women with bleeding disorders are more likely to undergo a hysterectomy and are more likely to have the operation at a younger age. While women with bleeding disorders are at risk for the same obstetrical and gynaecological problems that affect all women, women with bleeding disorders are disproportionately affected by conditions that manifest with bleeding. Optimal management involves the combined expertise of haemostasis experts and obstetrician-gynaecologists.

Introduction

While women are rarely affected by haemophilia, they are equally likely to have other bleeding disorders and when they do, are disproportionately affected as a result of the bleeding challenges of menstruation and childbirth. In the last 20 years, our understanding of the needs of women with bleeding disorders has increased and our ability to manage their needs has improved.

Menorrhagia and its relationship to bleeding disorders

It is now well-established that heavy menstrual bleeding, or menorrhagia, is the most common symptom that women with bleeding disorders experience. Over the last 20 years, data have been published about the prevalence of menorrhagia in women with bleeding disorders. These data are summarized in Table 1. It has been well-established that menorrhagia is more prevalent among women with bleeding disorders. In the last 13 years, it has also been established that bleeding disorders are more prevalent among women with menorrhagia. Among women with menorrhagia, the prevalence of von Willebrand disease (VWD) has been reported to be 5–20% [1–7] with an overall estimate of 13% [8] based on a systematic review. The prevalence of VWD and other bleeding disorders in women with menorrhagia is summarized in Table 2. There are limited data regarding the prevalence of bleeding disorders among adolescents with menorrhagia, but in the last 8 years, it has become apparent that they are at least as likely to have an underlying bleeding disorder as adult women with heavy menstrual bleeding. The prevalence of bleeding disorders in adolescents with menorrhagia (from studies of inpatients, outpatients and patients referred to a haemophilia treatment centre) are summarized in Table 3.

Table 1.   The prevalence of menorrhagia in women with bleeding disorders*.
Bleeding disorderPrevalence
  1. *Compared with the prevalence in women without bleeding disorders of 23–68.4% [75]

von Willebrand disease (VWD)32–100% (18)
Severe platelet dysfunction
 Bernard–Soulier syndrome51% (75)
 Glanzmann’s thrombasthenia98% (76)
Factor XI deficiency (of varying severity)59% (11)
Haemophilia carriage (with variably reduced levels of factor VIII or factor IX)10–57% (20, 37, 78, 79)
Rare factor deficiencies35–70% (46, 54, 80, 81)
Table 2.   The prevalence of bleeding disorders in women with menorrhagia.
Bleeding disorderPrevalence
von Willebrand disease (VWD)5–20% (1–7) Overall estimate 13%
Platelet dysfunction (depending on how defined)<1–47% (2, 4, 6, 7)
Factor XI deficiency<1–4% (2, 4, 6, 76)
Haemophilia carriage<1–4% (2, 4, 6, 76)
Rare factor deficiencies<1% (2, 4, 6, 76)
Table 3.   The prevalence of bleeding disorders in adolescents with menorrhagia.
Bleeding disorderPrevalence
von Willebrand disease (VWD)5–36% (83–89)
Platelet dysfunction (depending on how defined)2–44% (83, 85, 87–89)
Factor XI deficiency<1–4% (2, 4, 6, 82)
Low factor VIII8% (87)
Thrombocytopenia13–20% (85, 86)

Menorrhagia is defined as heavy menstrual bleeding that lasts for more than 7 days [9] or results in the loss of more than 80 mL of blood per menstrual cycle [9]. As measuring actual menstrual blood loss is not feasible in clinical practice, Higham et al. devised a pictorial blood assessment chart (PBAC) as an alternative [10]. In the investigators’ original study, women compared the degree of saturation of their pads and tampons with those depicted on a chart. Lightly stained pads or tampons obtained a score of 1, moderately stained pads or tampons a score of 5, and soaked pads or tampons a score of 20. The scores were summed and a total score of greater than 100 per cycle was associated with a menstrual blood loss of greater than 80 mL. A drawback of the use of the chart is that it must be completed prospectively and its results are not available at the time of an initial evaluation. Additionally, the validity of the chart remains uncertain [11,12]. Nonetheless, in the last 5 years, the PBAC has been used successfully to monitor response to treatment in studies of women with bleeding disorders [13,14].

In most situations, however, the practitioner must rely on a menstrual history and clinical impression to decide whether a woman has menorrhagia. Warner et al. attempted to assess the volume of blood loss by means of specific clinical features. In a logistic regression model, the variables that predicted menstrual blood loss of more than 80 mL were clots greater than a 50-pence in size (approximately one inch in diameter), low ferritin (according to the investigators’ laboratory reference) or changing a pad or tampon more than hourly (flooding) [15]. These are useful features that the practitioner can use to assess menorrhagia at the time of an initial visit.

Philipp et al. [16], also reported on the importance of flooding, not as confirmation of menorrhagia, but as a predictor of a bleeding disorder. The investigators administered a 12-page questionnaire of bleeding symptoms. Symptoms with high predictive values for laboratory haemostatic abnormalities were combined and used as single variables to calculate sensitivity, specificity and positive and negative predictive values to develop a short screening tool to identify females for testing and evaluation for a bleeding disorder. The screening tool was considered to be positive if one of the following four criteria was met:

  • 1 Duration of menses greater than or equal to 7 days and flooding or impairment of daily activities with most periods.
  • 2 History of treatment of anaemia.
  • 3 Family history of a diagnosed bleeding disorder.
  • 4 History of excessive bleeding with tooth extraction, delivery or miscarriage or surgery.

The screening tool alone had a sensitivity of 82% for bleeding disorders. Although the results would not be available at an initial visit, adding a pictorial blood assessment chart score >100 increased the sensitivity of the screening tool to 95%.

Other gynaecological bleeding

It has also been recognized that menorrhagia is not the only reproductive tract manifestation of a bleeding disorder. In a survey of 102 women with VWD conducted by the United States Centers for Disease Control and Prevention (CDC), the next most common reproductive tract abnormality that women with VWD reported after menorrhagia was a history of ovarian cysts (52% among cases vs. 22% among controls). Although ovulation is not normally accompanied by any significant amount of bleeding, in women with VWD or other bleeding disorders, ovulation can result in bleeding into the follicular sac, the peritoneum, the broad ligament and the retroperitoneum. In a case series of patients with VWD, Silwer found the incidence of haemorrhagic ovarian cysts in women to be 6.8% [17]. Haemorrhagic ovarian cysts have also been reported in women with afibrinogenemia, factor X deficiency, factor XIII deficiency, platelet defects or in women who are haemophilia carriers [18]. Acutely, surgery, tranexamic acid and clotting factor replacement have been used to manage haemorrhagic ovarian cysts [19–21]. Oral contraceptives, which suppress ovulation and may increase clotting factors, have been used to prevent recurrences [21–23].

In the same CDC survey, 30% of women with VWD reported a history of endometriosis compared to 13% of controls [24]. There are several possible reasons why females with VWD and other bleeding disorders would be more likely to be diagnosed with endometriosis, a painful condition in which endometrial tissue, which normally lines the uterus, implants in the abdomen outside of the uterus. Although there is disagreement regarding the aetiology of endometriosis, the prevailing theory is that it results from retrograde menstruation [25]. Heavy menstrual bleeding is a risk factor for retrograde menstruation and endometriosis [26]. Women with bleeding disorders have heavier menstrual bleeding, more retrograde menstruation and, possibly, more endometriosis. Alternatively, women with bleeding disorders may be more likely to experience symptomatic bleeding from endometriosis implants or have intra-abdominal bleeding that is misdiagnosed as endometriosis.

There is no evidence that women with bleeding disorders are more likely to develop fibroids (leiomyoma), polyps or endometrial hyperplasia (excessive growth of lining of the uterus), but in the same CDC survey of 102 women with VWD, 32% reported a history of fibroids vs. 17% of controls, 10% reported a history of endometrial hyperplasia vs. 1% of controls and 8% reported a history of polyps vs. 1% of controls [24]. The development of one of these conditions may unmask a previously subclinical bleeding tendency and, in a woman with a bleeding disorder, may cause problematic bleeding. It is clear from these data that the presence of gynaecological pathology does not exclude the existence of a bleeding disorder.

Management of menorrhagia in women with bleeding disorders

As menorrhagia may be a sign of a gynaecological problem other than a bleeding disorder, a full gynaecological evaluation is required prior to treatment of menorrhagia [9,27]. With the exception of non-steroidal anti-inflammatory drugs, which may affect platelet function and systemic haemostasis [28] and are not generally prescribed for patients with bleeding disorders [29], any gynaecological treatment that will reduce heavy menstrual bleeding may be appropriate depending on a woman’s age, gynaecological conditions and reproductive plans.

Oral contraceptives have been found to reduce menstrual blood loss in women with VWD [30], and possibly increase von Willebrand factor and factor VIII levels [31–33]. Oral contraceptives have been used to reduce menstrual blood loss in women with other bleeding disorders as well. Although there are no accumulated data regarding the use of other hormonal therapies in women with bleeding disorders, there is no reason to believe that other combined hormonal contraceptives such as patches and rings would not also be effective in reducing menstrual blood loss.

What has not been well studied are the benefits of one formulation or dosing strategy compared with another in the reduction of heavy menstrual bleeding, especially in women with bleeding disorders. In one randomized trial of women without bleeding disorders who were taking continuous combined hormonal contraceptive pills, the addition of 10 mcg of ethinyl estradiol to a 20 mcg ethinyl oestradiol pill containing levonorgestrel or norethindrone acetate did not improve bleeding patterns. However, preparations containing 1 mg of norethindrone acetate resulted in fewer days of bleeding or spotting compared with preparations containing 100 mcg levonorgestrel [34].

In the last 5 years, the levonorgestrel intrauterine device (IUD) (Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA) has been shown to be effective at reducing menstrual blood loss in women with bleeding disorders [13]. Other progestin-only contraceptives, such as Depo-Provera® (medroxyprogesterone acetate) injections (Pharmacia & Upjohn Company, NY, NY, USA), progestin-only pills, and the Implanon® implant (Organon, Roseland, NJ, USA) should also reduce endometrial proliferation and reduce menstrual blood loss. Medroxyprogesterone acetate is now available in a subcutaneous formulation, depo-subQ provera 104™, providing an alternative to the need for intramuscular injection. Insertion of the Implanon® implant could also cause bleeding in a woman with a bleeding disorder and might require pretreatment with a haemostatic agent.

Women who have had a proper gynaecological evaluation and fail hormonal management or desire pregnancy should be considered for haemostatic therapies, which have been demonstrated to be effective in controlling menorrhagia in women with bleeding disorders. These haemostatic therapies include DDAVP (1-desamino-8-D-arginine vasopressin), antifibrinolytic medications (aminocaproic acid and tranexamic acid) and clotting factor concentrates. A recent multi-site prospective cross-over study demonstrated that both DDAVP and tranexamic acid reduce menstrual blood flow in women with bleeding disorders. With DDAVP, there was a decrease in the PBAC score of −66.0 vs. −107.8 with tranexamic acid. This difference was statistically significant, although both treatments improved quality of life [14].

Dilation and curettage (D & C) has historically been used to both diagnose and treat heavy menstrual bleeding, but in the last 10 years, D & C has largely been replaced by less invasive options such as ultrasound, endometrial biopsy and hysteroscopy [35,36]. These alternatives are more appropriate for women with bleeding disorders, as D & C may actually increase bleeding in women with bleeding disorders [20,37].

Moreover, in the last 10 years, women with bleeding disorders who have completed childbearing have had a less-invasive alternative to hysterectomy for definitive management of their heavy menstrual bleeding. Endometrial ablation, which requires no incisions and does not carry the same risks as hysterectomy, has been demonstrated to be as effective in reducing menstrual blood loss in women with bleeding disorders as in other women [38]. Nonetheless, a recent meta-analysis of six randomized clinical trials suggested that the levonorgestrel IUD is as effective as endometrial ablation in reducing menstrual blood loss [39] and a previous small randomized trial found the levonorgestrel IUD to be nearly as effective as endometrial ablation in reducing menstrual blood loss [40]. Therefore, the levonorgestrel IUD should still be considered for the management of menorrhagia, even for women who have completed childbearing.

Women who have completed their childbearing, particularly those who have failed medical management or endometrial ablation, are candidates for hysterectomy. Because menorrhagia is often the primary symptom that women with bleeding disorders experience, hysterectomy does offer the possibility for significant improvement in quality of life and is a safer procedure than it was decades ago. Thirty-six years ago when Silwer et al. published the results of a comparison of the complications of hysterectomy in 18 women with VWD vs. 50 controls, 50% of the women with VWD received transfusion, but so did 30% of the controls [17]. In a recent study of United States hospital discharge data, only 7% of women with VWD received transfusions compared to 2% of controls [41]. Furthermore, in the last 20 years, the availability of recombinant or plasma-derived/virally inactivated clotting factor concentrates has reduced the chance of viral transmission with factor replacement.

There are few data on management of acute, severe menorrhagia, particularly in the adolescent or woman with a bleeding disorder. In November, 2009, a consensus conference sponsored by CSL Behring was convened specifically to address this issue. A full report will be published in the future, but there was consensus that balloon tamponade, hormonal therapy (oestrogen) and antifibrinolytic treatment should be instituted while replacing clotting factor or platelets as indicated.

Pregnancy and childbirth

It is recognized that normal pregnancy is accompanied by increased concentrations of various clotting factors. Despite improved haemostasis, however, women with bleeding disorders often do not achieve the same levels of clotting factors as other women and, therefore, are at an increased risk of bleeding complications with pregnancy.

In the last 20 years, there have been several case reports and case series documenting the profoundly increased risk of miscarriage and placental abruption resulting in foetal loss or premature delivery in women with deficiency of fibrinogen [42–51], or factor XIII [52–55] but whether there is an increased risk of miscarriage in women with other bleeding disorders is not clear [18]. Clotting factor replacement is used to reduce the risk of miscarriage, foetal loss and premature delivery in women with deficiency of fibrinogen [43–45,47–51] and factor XIII [52–55], but whether any therapy is necessary or available to prevent miscarriage or foetal loss in women with other bleeding disorders has not been reported.

Despite the primary role of uterine contractility in controlling postpartum blood loss, women with bleeding disorders are at an increased risk of postpartum haemorrhage. There are multiple case series documenting the incidence of postpartum haemorrhage in women with bleeding disorders [18] and four case-control studies comparing women with VWD and controls. The most recent, from a United States discharge database, found that 6% of pregnancies in women with VWD were complicated by postpartum haemorrhage compared to 4% of controls [56]. Perineal haematoma, a rare complication of vaginal birth, occurs with some frequency in women with bleeding disorders [18] and contributes to the increased incidence of postpartum haemorrhage.

In women with bleeding disorders, haemorrhage, when it does occur, has frequently been reported to occur more than 2–3 weeks postpartum. In normal pregnancies, the median duration of bleeding after delivery is 21–27 days [57–59]. Clotting factors, which are elevated during pregnancy, return to pre-pregnancy levels within 14–21 days [60]. Because women generally continue to bleed after clotting factors have returned to pre-pregnancy levels, women with bleeding disorders may be particularly vulnerable to delayed or secondary postpartum haemorrhage during this time. While delayed or secondary postpartum haemorrhage is rare, occurring after fewer than 1% of deliveries [61,62], delayed postpartum haemorrhage has been reported in 20–25% of women with VWD [63,64], 2–11% of haemophilia carriers [20,65] and 24% of women with factor XI deficiency [64].

Management of pregnancy and childbirth

Ideally, planning for pregnancy begins before conception. Prior to conception, or during pregnancy, women should be offered the opportunity to speak with a genetic counsellor regarding the inheritance of their bleeding disorder [66] and with a paediatric haematologist regarding the care of a potentially affected child. Women and their families should be apprised of the full range of prenatal diagnostic options that exist (chorionic villus sampling, amniocentesis, foetal sex determination by ultrasound or foetal sex determination through maternal plasma, if available) as well as the option of pre-implantation diagnosis, which has led to the successful live birth of at least one child [67].

The management of childbirth will depend on the needs of the mother and her potentially affected infant at the time of delivery. Women at risk for severe bleeding should be referred for prenatal care and delivery to a centre where, in addition to specialists in high-risk obstetrics, there is a haemophilia treatment centre or a haematologist with expertise in haemostasis. Laboratory, pharmacy and blood bank support is essential. Prior to delivery, all women with bleeding disorders should have the opportunity to meet with an anesthetist. There is no consensus on the factor levels that are safe for regional anaesthesia, but if levels are at least 50%, and the rest of the coagulation studies are normal, regional anaesthesia may be considered safe.

Prior to any invasive procedure such as chorionic villus sampling, amniocentesis or cervical cerclage, women at risk for severe bleeding should receive prophylaxis. DDAVP, if required during pregnancy, is generally thought to be safe for mother and foetus [68,69]. At the time of childbirth, DDAVP must be used with caution, if at all. DDAVP causes fluid retention and patients receiving DDAVP should be fluid restricted, yet women commonly receive 1–2 L or more of fluid at the time of a vaginal delivery and 2–3 L or more at the time of caesarean delivery. To prevent postpartum haemorrhage after delivery, women routinely receive oxytocin, which also causes fluid retention. Administration of DDAVP, combined with litres of fluids and oxytocin, may result in life-threatening hyponatraemia [70]. A single dose of DDAVP immediately prior to epidural catheter placement in labour, however, has not been associated with adverse events [71].

Among the published series of VWD in pregnancy there are multiple cases of postpartum haemorrhage that occurred despite prophylaxis [18]. In a review of published cases of women with VWD who experienced postpartum haemorrhage, Roque et al. [72] determined that the average time of haemorrhage was 15.7 ± 5.2 days after delivery. The implication is that women with bleeding disorders may require more frequent evaluation. Thus, weekly contact is suggested during the postpartum period [73]. Prophylaxis, when indicated, may be required for two or more weeks. More data are required to determine optimal length of prophylaxis.

Guidelines

Data on the management of women with bleeding disorders are hampered by a lack of randomized trials, case-control studies or even large case series. No one centre sees a large number of patients. Severe bleeding disorders are rare and women with milder disease may not come to the attention of a haemostasis centre or even be diagnosed. Funding for studies is limited. In the absence of strong evidence to direct practice, government agencies and haemophilia organizations have developed consensus guidelines. There are at least nine sets of guidelines published by the government agencies or haemophilia organizations that specifically address women with bleeding disorders. The guidelines were reviewed and summarized in 2009 and found to be remarkably congruent [74]. The good news is that there is consensus regarding many of the issues pertaining to the management of women with bleeding disorders.

Ancillary