Management of carriers and babies with haemophilia

Care and treatment of carriers

Treatment with factor VIII (FVIII) and factor IX (FIX) concentrates during the last decades has dramatically improved morbidity and mortality in haemophilia A and B. In Sweden in 1960, for example, the mean age of death for a person with haemophilia was 23 years but today, with early prophylactic treatment, life expectancy is almost normal [1]. As a consequence, males affected with haemophilia will reproduce and have daughters who are genetically obligate carriers, i.e. the number of carrier women has increased in the population [2]. Accurate carrier diagnosis is now possible by DNA analyses and many female members in families with haemophilia A or B have been investigated for carriership. Furthermore, many pregnant haemophilia carriers request prenatal diagnosis although they do not want to interrupt pregnancy but rather be psychologically prepared for having a child with haemophilia. Health care systems will thus be challenged with an increasing number of well-informed carriers with questions and thoughts on applications of genetic information and management of pregnancy and delivery.

Many haemophilia centres are expanding their services to deal with rare bleeding disorders other than haemophilia and von Willebrand disorder bleeding disorders in women. Women who are carriers of the genes for FVIII or FIX deficiency may have plasma factor levels in the reference ranges but up to 80% may have levels below these. Some carriers will have plasma factor levels in the mild haemophilia range (5–30%, reference range 50–150%) and require haemostatic support during surgical challenges and delivery. They are often called ‘symptomatic carriers’. It is important that obligate carriers and other females at ‘risk’ of being carriers have their factor levels (phenotype) measured. It is also very important that they and their families understand that having a normal factor level does not exclude carriage of the gene, which must be tested by genetic analysis. This may not be appreciated by clinicians outside a specialized bleeding disorder centre and referral to such a centre for information, support and access to phenotypic and genotypic studies is necessary.

Women defined by bleeding experiences and/or phenotypic testing to be symptomatic carriers require ongoing support from bleeding disorder specialists at times of surgical procedures and delivery.

Timing of testing

Testing phenotypic status is important before a potential carrier has surgery or with trauma to assess the need for haemostatic support. A low factor level will betray but not define genetically carrier status. Genetic testing is performed as part of a supportive clinical encounter with informed clinicians, genetic counsellors and excellent laboratory technologists.

Pregnancy is not the best time to start discussions and testing

Wherever possible women should be tested and aware of their genetic status and reproductive options before this time. Testing for FVIII levels, where possible, and von Willebrand factor (VWF) levels, where appropriate, should be performed before starting hormone therapies for excessive menstrual bleeding. Another advantage of early phenotypic testing is to be able to advise against the prescription of drugs which may aggravate bleeding (such as non-steroidal anti-inflammatory agents).

There are many strategies available to carriers to reduce menorrhagia which is the most common symptom of bleeding. These include the use of oral and intra-uterine hormone therapies to reduce endometrial proliferation, desmopressin (DDAVP) in women with von Willebrand disorder and FVIII deficiency and cyclokapron.

Genetic testing

Phenotypic testing does not reveal carrier status in many people tested. The timing of genetic testing needs careful consideration taking into account age, psychological and cultural issues. Genetic information permits reproductive choices where available. It is not to be used for ‘eugenic’ purposes. Carriers and their familiars should have assured comfort and access to genetic counsellors and other clinicians in preparation for testing and receiving results.

Management of pregnancy

Management of a carrier mother during pregnancy should be co-ordinated between her obstetrician, anaesthetist and haematologist. Where possible, foetal gender should be determined, traditionally by ultrasound examination, to prepare parents and obstetricians for planning of the safe delivery of a boy who potentially has haemophilia [3]. Plasma factor levels of VWF and FVIII may rise sufficiently during pregnancy to permit safe haemostasis without exogenous haemostatic support for epidural analgesia and other possible interventions at delivery. FIX levels do not rise in pregnancy. It is prudent to measure VWF and/or FVIII levels at 32–34 weeks of pregnancy. Where the levels are less than 50% and more particularly less than 30% of the reference ranges, DDAVP or FVIII concentrates may be used.

There is a little published data on the use of DDAVP in pregnancy women who are symptomatic carriers of FVIII deficiency. The prescribers’ information advises the drug is contra-indicated with lactation and recommends precaution with pregnancy [4]. Mannucci reported the use of DDAVP in 27 carriers of FVIII deficiency in the first and second trimesters of pregnancy without adverse events in mothers or foetuses that progresses to term [5].

Sanchez-Luceros et al. have reported recently on the antenatal use of DDAVP in 54 women with a bleeding history and low VWF levels, including use to support epidural blockage and delivery. No hyponatraemia or thromboembolic complications were seen and babies exposed to DDAVP in all trimesters were healthy. The potential for maternal and/or neonatal hyponatraemia was reduced by prudent fluid management and DDAVP did not have a uterine muscle stimulatory effect [6].

The use of DDAVP in early puerperium to counteract the postdelivery fall in VWF/FVIII levels has again been advocated in recent guidelines on the Obstetric and Gynaecological Management of Women with Inherited Bleeding Disorders from a Taskforce of UK Haemophilia Centre Doctors’ Organization [7] (this article includes a list of helpful websites for the interested reader).

Risks to the neonate at delivery

The risk of intracranial haemorrhage (ICH) is the most serious threat during delivery of a haemophilia boy, the second most serious being extracranial haemorrhage (ECH). There is no universally accepted figure regarding the frequency of ICH in neonates, since it is dependent on general issues such as the health care status of the population studied; maternal issues such as knowledge of carriership and the status of the foetus; obstetric issues such as guidelines for the management of carrier women and mode of delivery; and finally surveillance and routine management of the newborn with known or suspected haemophilia.

Kulkarni and Lusher reviewed the literature in 1999 with respect to newborns with haemophilia [8]. Their Medline search resulted in 33 publications describing 102 newborns with haemophilia and cranial bleeds. The cumulative incidence of ICH and ECH was calculated to be 3.6% based on the five studies that reported a denominator for the newborn population. There were a total of 109 episodes; 65% ICH, 35% ECH and a 6% combination of both.

Ljung et al. made a survey of every child with severe or moderate haemophilia A or B, born in Sweden during the period 1970–1999 [9]. All 117 case records were surveyed for mode of delivery and complications. Neonatal complications were: subgaleal or cephalic haematoma (n = 12), ICH (n = 4), umbilical bleeding (n = 4), haematuria (n = 1), retro-orbital bleeding (n = 1) and abnormal bleeding after surgery, injection or venepuncture (n = 28). Of the 4/117 (3.5%) children with ICH, all were sporadic cases of haemophilia, one was a premature birth by caesarean section in the 27th week, one was delivered by vacuum extraction and the remaining two vaginally. In these four cases, there were no sequelae or only minor ones.

A French multi-centre survey of ICH conducted from 1991 to 2001 revealed 123 episodes in 106 patients, of whom 10 (one-third of the cases in children) occurred in the neonatal period. However, the denominator population is not presented [10]. In a retrospective multi-centre study of ICH performed in Canada during the interim 1984–2000, 37% (n = 172) occurred during the first week of life (3.7% of the total) [11].

In summary, the most recent studies show a rather uniform pattern that 3.5–4.0% of all haemophilia boys born in countries with a good standard of health care will suffer from ICH during the neonatal period. This is considerably higher than expected in the normal population, where published reference figures suggest an incidence of one ICH of 860 deliveries with vacuum extraction, 1/907 caesareans during labour, 1/1900 in spontaneous delivery and 1/2750 in elective caesareans [12].

Recommended mode of delivery

Different opinions exist regarding the mode of delivery for pregnant haemophilia carriers. It seems that in most centres, normal vaginal delivery is the recommended mode of delivery [8,9]. Caesarean delivery does not eliminate the risk of cranial haemorrhage in newborns with haemophilia [8]. A study by MacLean et al., in the Netherlands, showed that instrumental delivery was twice as frequent in women who were unaware of their carrier status compared with those who were aware, 16% vs. 8%, respectively [13]. The frequency of caesarean section was 22% in the subgroup of known carriers vs. 8% in the other group. The caesarean section rate was 47% in known carriers (and most of them with knowledge that the foetus was male) in the recently published series from Royal Free Hospital (n = 65) [14]. This demonstrates a more cautious approach at delivery in carriers. However, in a survey of obstetricians in the USA 1999, although there was a high percentage of non-responders, only 11% recommended routine caesarean [8].

All series presented are biased against complications because they include birth of children with sporadic haemophilia, i.e. the disorder was not known in the family before [15]. Since approximately 50% of the newborns with haemophilia are sporadic cases, it is difficult to get an accurate figure on the risk of complications in a group with known carriers when this knowledge (and even more if the sex or a prenatal diagnosis of haemophilia of the foetus is known [3]) will impact the way a normal vaginal delivery is performed.

Chi et al. have reviewed their experience of complications, management and outcomes of pregnancy in carriers of haemophilia over a 10-year period following the introduction of a multidisciplinary management guideline as summarized in Table 1 [14].

It is important that mothers have adequate levels of plasma factors for epidural injections, episiotomy and caesarean section. Procedures, in particular Ventousse suction extraction and high-forceps deliveries, should be avoided because of the risk of neonatal intracranial and other haemorrhages.

Management plans should be agreed and written in advance by all consultants. The plan emphasizes the safe and careful delivery of a boy with potential haemophilia, with reminders as to the avoidance of vacuum extraction, mid-cavity forceps delivery and prolonged labour. The babe should not have scalp electrode placement or scalp blood sampling wherever possible, nor deep intramuscular injections after birth until gender and coagulation status are confirmed.

In summary, based on the current literature, one may conclude that the risk of serious bleeding in the neonate in conjunction with normal vaginal delivery is small when appropriate precautions are taken [8,9,16]. Another important issue, out of the scope of this paper, is the management of the newborn with known or potential haemophilia [17].

Pre-implantation genetic diagnosis

This form of diagnosis is the most attractive to many couples at risk of having a child with haemophilia and technology is becoming more widely available. Pre-implantation genetic diagnosis (PGD) is a very early form of prenatal diagnosis. It combines molecular and cyto-genetics with assisted conception techniques such as in-vitro fertilization (IVF) to allow identification of abnormalities in embryos prior to implantation. Couples at risk of having children with serious genetic conditions such as haemophilia can have unaffected embryos transferred to the uterus, eliminating the need for prenatal diagnosis and termination of pregnancy. Handyside and his team at the Hammersmith Hospital London first performed PGD successfully in 1990 [18]. Prior to the introduction of PGD, couples at risk of having children with haemophilia faced reproductive uncertainty and anxiety. They could consider invasive prenatal diagnosis (with a 1% risk of losing an unaffected pregnancy) and then consider the difficult and traumatic decision of terminating or continuing an affected pregnancy: a form of reproductive roulette. Some couples chose simply not to have children, whilst others chose to adopt. Gamete donation (in the case of haemophilia, this meant egg donation) was another option. In the programme at Imperial College, London, three women had been so traumatized by caring for children who subsequently died that they chose to be sterilized [19]. The aim of developing a PGD programme was to extend the range of reproductive options to these couples: hopefully reducing anxiety, avoiding invasive prenatal diagnosis, eliminating the need for termination of pregnancy and avoiding the birth of children affected with haemophilia.

Methods of PGD for haemophilia

The analytic techniques of determining haemophilia status from 6 pg of DNA in one embryonic cell need to be accurate and reliable; diagnosis from a single cell must be representative of the entire embryo; there should be minimal risk of impairing foetal development and finally there must be an acceptable chance of pregnancy following embryo replacement. There are three potential methods for the PGD of haemophilia: sperm sorting, egg biopsy (or more accurately polar body biopsy) and embryo biopsy. Previous attempts at sperm sorting have included albumin density gradients, modified swim-up, discontinuous percoll, quinacrine staining, however, enrichment is insufficient for clinical use. Recently, a technique using a fluorescent dye that binds to DNA has been evaluated [20]. X- and Y-bearing sperm are then separated using a flow cytometer. The sperm can then be used for simple insemination or in assisted conception treatments such as IVF. There are some concerns about the technique resulting in subtle damage to DNA, and as a result it is currently only being used within the context of a clinical trial [21].

The haemophilia gene could also be detected in the egg or oocyte. During meiosis, a polar body is extruded from the oocyte that can be biopsied and analyzed without harming the oocyte. If a woman is known to be a haemophilia carrier, and the polar body is found to contain the mutation in the gene, then it can be inferred that the egg will have a normal copy of the gene, and vice versa. Proponents of this technique argue that it is more ethically acceptable to biopsy (and potentially discard) eggs rather than embryos. Although PGD for haemophilia has been performed this way [22], there have been concerns regarding ‘crossing over’ of genetic material during meiosis requiring subsequent second polar body and even blastomere biopsy.

The majority of centres offering PGD for haemophilia do so using biopsy of the embryo at the cleavage stage [23]. This requires the couple undergoing controlled ovarian stimulation using gonadotropin injections. Eggs are then collected surgically by the transvaginal route and inseminated or injected with sperm. Embryos are then cultured for three-day postfertilization, when they should be eight cells in size. Embryos are then biopsied and one or two cells are removed for analysis. Initially, the patients carrying haemophilia were offered sexing of their embryos with the transfer of only female embryos. Embryo sexing was first performed using a PCR approach [24] but this was stopped following misdiagnoses because of allele drop-out (non-amplification of the Y locus). Fluorescent in-situ hybridization was then tried for the X and Y chromosome (usually with control chromosomal markers) with considerable success [25], and this became the mainstay technique for haemophilia PGD.

Embryo sexing is not an ideal approach for the PGD of haemophilia: all male embryos were discarded (50% of which would have been normal). A mutation specific approach would mean that 75% of embryos would be available for selection (increasing pregnancy success rates because more better quality embryos would be available) and offering the chance of the couple having male children. An indirect approach has been successful for a variety of X-linked disease using co-amplification of polymorphic markers [26]. The team at Hammersmith IVF developed a simple restriction enzyme approach in couples who had single base pair substitutions in the FVIII gene. This led to the livebirth of an unaffected baby girl [27]. This technique was extremely labour intensive, and not widely applicable to many couples who carried different mutations within the gene. As fluorescent PCR and in particular, the speed of sequencing advanced, techniques with a more broad application have been developed [28]. These techniques also increase the safety of single cell PGD because multiplex fluorescent amplification of four short tandem repeats was adapted to a single cell pre-amplification in order to rule out contamination and allele drop-out, and for confirmatory indirect diagnosis.

Perhaps one of the most exciting developments in the molecular diagnosis of single embryonic cells has been pre-implantation genetic haplotyping [29]. Here, multiple displacement amplification is used as the initial PCR step followed by genotyping using multiple polymorphic markers (up to 57). The authors conclude that this represents a paradigm shift in embryo diagnosis, as one panel of markers can be used for all carriers of the same monogenic disease, bypassing the need for development of mutation-specific tests, and widening the scope and availability of pre-implantation genetic testing. This could mean that PGD could become much more readily available to carriers of haemophilia and possibly cheaper.

Ethical challenges

Pre-implantation genetic diagnosis has always attracted a wide ethical commentary. The opinions are usually divided into the deontological position which proposes that the creation, selection and destruction of human embryos is fundamentally wrong and cannot be justified whatever the proposed beneficial outcome; vs. a more utilitarian view which suggests that in a carefully regulated environment where only serious genetic disease is considered that the prevention of life-threatening disease does justify this use of embryos. The consequentialist argument or ‘slippery slope’ argument is also frequently used, particularly towards a free market version of eugenics. Supporters of PGD state that the basic tenets of medical ethics: beneficence, non-malficience, autonomy, justice are all satisfied in their practice. There is wide variation in the regulation of PGD internationally, in some states decisions are left up to the commissioning couple and their physician, whilst in others there is considerable legislation and licensing conditions enforced on clinics, whilst in some countries the practice remains illegal [30]. Many clinicians believe that such external regulation can inhibit research. However, experience in the UK over the last 18 years is that legislation can be permissive and need not always be restrictive [31]. It can also help reassure the public that cutting-edge scientific practice is being performed with ethical integrity.

The future of PGD for haemophilia

It is likely that new technical approaches such as haplotyping will allow a universal specific diagnosis without the need for sexing. However, this remains an expensive, high-tech option, and the demand will probably remain small. If sperm sorting is proved to be safe, it offers a low-tech, cheaper and more accessible option to the majority of the world’s population. However, it needs to be remembered that it only results in a female livebirth 90% of the time. As a general principal, the future involves increasing safety, accuracy and pregnancy rates. Good practice guidelines for clinical, embryological and laboratory practice have recently been published [32]. Of equal importance is improving the profile of PGD, leading to a wider acceptance amongst the public and professions. This can only be performed by dealing with the pressure to balance parental autonomy with ethical concerns of society.