Description of the condition
Bleeding disorders can pose significant challenges during pregnancy and childbirth. Pregnancy is a hypercoagulable state, which means that the majority of bleeding disorders are not associated with a significant bleeding risk. However, some of the disorders will not be corrected by the hypercoagulable state induced by pregnancy, or this state will quickly wane post-delivery, thus requiring specific measures to prevent maternal and foetal bleeding complications especially during childbirth. Von Willebrand disease (VWD), haemophilia A and B, factor XI and factor VII deficiency account for almost 90% of the inherited bleeding disorders, whereas deficiency in fibrinogen, prothrombin factor, factor V, factor X, factor XIII are relatively rare (Chi 2007). Idiopathic thrombocytopenic purpura (ITP) is a commonly acquired bleeding disorder during pregnancy and accounts for 3% to 4% of the cases of thrombocytopenia detected at this time (Gill 2000); whereas HELLP (haemolysis, elevated liver enzymes, and low platelet count) syndrome is a rarer acquired bleeding disorder during pregnancy, occurring in approximately 1 to 2 per 1000 pregnancies (Geary 1997).
- Von Willebrand disease is caused by a quantitative or qualitative defect in Von Willebrand factor (VWF). An increase in VWF (with levels in blood plasma up to normal plasma levels) is often seen during pregnancy in women with VWD (except type 3), but intrapatient variability is very wide (Conti 1986). A drastic fall in VWF following delivery in women with VWD causes these women to be prone to primary and secondary postpartum haemorrhage (PPH).
- Congenital haemophilias are X-linked recessive bleeding disorders that result from deficiencies of the coagulation factors VIII and IX. Most female carriers of haemophilia have levels of factor VIII (or IX) within the normal range and are therefore protected against significant bleeding problems in day-to-day life (Giangrande 2009). Pregnant carriers of haemophilia A usually have low levels of factor VIII at the beginning of pregnancy, but have almost normal levels of factor VIII by the third trimester. In normal pregnancy, factor IX levels do not rise significantly (Stirling 1984). However, those with levels of factor VIII (or IX) under 50% (less than 0.50 IU/mL) are at increased risk of bleeding when facing haemostatic challenges (Lusher 1978). Female carriers of haemophilia A are at an increased risk of haemorrhagic complications, both early and late PPH, because of the rapid fall in the increased pregnancy-induced maternal clotting factor levels (FVIII) after delivery. The incidence of early and late PPH is increased among haemophilia A carriers (22% and 11%, respectively) (Kadir 1997) compared with the general population (5% and 0.7%, respectively) (Lee 1981). Risk of head bleeding in babies (cephalhaematoma, intracranial haemorrhage) is 3%, 64% and 15% following normal delivery, vacuum extraction and caesarean section (c-section) respectively (Ljung 1994)
- Factor XI deficiency (haemophilia C) and factor VII deficiency are very rare inherited bleeding disorders in the general population but factor XI deficiency is very common in the Ashkenazi Jewish population (Bolton-Maggs 1988). Homozygous and compound heterozygous individuals with factor XI and factor VII deficiency develop haemorrhagic manifestations; heterozygous individuals are usually asymptomatic. Clinical prototypes vary from mild to severe and do not correlate with factor XI and VII levels. The levels of factor XI and factor VII do not change during pregnancy and there is a poor correlation between the level of factor XI and bleeding tendency (Bolton-Maggs 1988). For both factor XI and VII deficiencies, bleeding tendency is likely to be associated with levels of factor less than 15%, and varies in the same individual following different haemostatic challenges. Therefore, the unpredictable nature of these diseases makes their management during pregnancy and childbirth difficult.
- Congenital platelet dysfunction disorders, being autosomal recessive disorders, are rare. Defects are in the platelet GPIb complex (Bernard-Soulier Syndrome) or the GPIIb-IIIa complex (Glanzmann thrombasthenia) or in the abnormal secretion and thromboxane synthesis. Bleeding tendencies are severe in Bernard-Soulier Syndrome and Glanzmann thrombasthenia, and need platelet transfusion. Bleeding is mild in platelet dysfunction disorders, which are due to abnormal secretion and thromboxane synthesis, but can be life-threatening following surgery or trauma (George 1990; Peng 1991).
- Idiopathic thrombophilic purpura is a common cause of acquired bleeding disorder in pregnancy and occurs in 1000 to 2000 deliveries (Burrows 1990). The risk of spontaneous bleeding at birth in women with ITP is low and occurs particularly if the platelet count decreases to less than 20,000/µL (Webert 2003). The frequency of neonatal alloimmunisation thrombocytopenia (NAIT) is estimated at 1 to 2 cases per 1000 deliveries (Blanchette 1990). This disorder develops in foetal life, with 25% to 50% of foetal intracranial haemorrhages detectable on prenatal ultrasound prior to the onset of labour (Herman 1986).
- The HELLP syndrome occurs in 10% to 20% of cases with severe pre-eclampsia (Geary 1997). Maternal mortality in pregnancy associated with HELLP syndrome is 1.1% (Sibai 1993) and the perinatal mortality is 34% before 32 weeks gestation, and 8% after this (Gul 2005). More common and serious maternal complications are placenta abruptio, disseminated intravascular coagulation (DIC) and subsequent severe PPH; whereas, prematurity, intrauterine growth restriction (IUGR) and abruptio placenta are the leading causes of neonatal death (Magann 1999). Hepatic rupture has a perinatal mortality that can reach 80% (Mihu 2007), the majority of which are observed during the antenatal period; whereas, about 30% of HELLP syndromes develop after birth, the majority within the first 48 hours with the time of onset ranging from a few hours to seven days following delivery (Barton 2004). Immediate delivery following stabilization of the maternal clinical condition is the primary choice of management in HELLP syndrome.
Description of the intervention
with a level less than 15% (i.e., less than 15% factor VII in factor VII deficiency;less than 15% factor XI in factor XI deficiency).
Pregnancy and childbirth management in women with bleeding disorders is a multidisciplinary approach consisting of obstetricians, haematologists, anaesthesiologists and neonatologists. Delivery is conducted in centres where resources for laboratory testing and clotting factor treatment are readily available. Relevant coagulation factor level is usually not measured at delivery, but is estimated during the third trimester. A deficient coagulation level has to be corrected to a safe level spontaneously or by supplementation in order to prevent bleeding during childbirth. Optimal management of delivery has to take into account the health needs and goals for both the mother and the newborn.
Delivery should be achieved by the least traumatic method and early recourse to c-section can be considered to minimise the risk of neonatal bleeding complications. Caesarean section is better for the foetus and not riskier for the mother, although the latter has not been proven in this context. Prolonged labour should be avoided whenever possible. Maternal genital and perineal trauma is minimised in order to reduce the risk of excessive bleeding at delivery (Kadir 1998; McMahon 2001).
Vaginal delivery is a safe mode of delivery in women with bleeding disorders and c-section section is mainly due to obstetric reasons (Lee 2006). Special attention to haemostasis is required to minimise bleeding during c-section. Administration of neuraxial or general anaesthesia are safe methods for labour and delivery. Risk of spinal epidural haematoma following administration of neuraxial anaesthesia may be increased with potential for permanent neurological injury, which is rare (Vandermeulen 1994). The decision for neuraxial anaesthesia is individualised and the procedure is better performed after optimum level of deficient clotting factor is achieved. When there is concern about spinal epidural haematoma, patient-controlled analgesia with fentanyl is an alternative option (Campbell 2003). Caesarean section under general anaesthesia is considered in cases where neuraxial analgesia is contraindicated. Instrumental delivery with low forceps is permissible, delivery by ventouse extraction or mid-cavity or rotational forceps delivery is not recommended and episiotomies and foetal scalp electrodes are avoided. Prophylactic replacement therapy is considered if maintaining the haemostatic level for at least five to seven days following either vaginal or c-section to prevent PPH is advisable. Prophylactic administration of a uterotonic drug (oxytocin, ergometrine, misoprostol, carboprost) along with active management of the third stage of labour is considered to prevent PPH.
Bleeding complications during delivery are 1% to 4% (intracranial haemorrhage, cephalhaematoma) in affected infants of inherited bleeding disorders and ITP, and commonly observed with traumatic delivery. However, these risks are not decreased by c-section (Fujimura 2002). Newborns affected by congenital coagulopathies are supplemented with oral instead than parenteral vitamin K. Similarly, for immunisation of affected newborn and neonates the intradermal or subcutaneous route is preferred.
How the intervention might work
Vaginal delivery is considered safer compared to c-section in women with bleeding disorders. It avoids surgical morbidity and anaesthesia-related complications. The safest method of delivery for foetuses at risk is controversial. Lack of data from the literature in relation to foetuses with rare bleeding disorders and knowledge is extrapolated from experience of newborns with common bleeding disorders such as haemophilia (Ljung 1994).The available data indicate that the risk of serious bleeding during normal vaginal delivery is small and delivery by c-section is not expected eliminate this risk (Ljung 1994). Low forceps delivery is considered less traumatic compared to c-section when the head is deeply engaged in the pelvis and easy outlet delivery is anticipated (Kadir 1998; McMahon 2001).
Why it is important to do this review
At present, there is ongoing debate about the best way to deliver in carriers of foetuses with, or at risk of, congenital bleeding disorders. The evidence from the comparison among vaginal and surgical delivery is scanty and there is equipoise which would justify controlled clinical trials being undertaken. This review will investigate which is the safe and more effective mode of delivery in women with bleeding disorders taking into account effects on both the mother and the foetus.
To assess the optimal mode of delivery in women with, or carriers of, bleeding disorders.
Criteria for considering studies for this review
Types of studies
Randomised control studies and all types of controlled clinical trials.
Types of participants
Pregnant women with, or carriers of, any type of bleeding disorders.
Types of interventions
Vaginal compared to c-section delivery.
Types of outcome measures
- Maternal death
- Post partum haemorrhage (the trial authors' definitions to be accepted when using or including one of the following definitions)
- blood loss (a haemoglobin drop ≥ 2 g/dL or a haematocrit drop ≥ 10% (or both))
- blood component transfusion (≥ 2 units of packed cell transfusion, fresh frozen plasma, factor transfusion)
- Infant death
- Severe maternal morbidity
- major surgery
- vital organ failure
- intensive care unit (ICU) admission
- wound infection
- wound haematoma
- low genital tract haematoma
- Instrumental delivery
- Foetal morbidity
- head bleeding (e.g. cephalhaematoma, intracranial haemorrhage)
- Other anaesthesia complications
- spinal haematoma
- epidural haematoma
- Length of the stay
Search methods for identification of studies
We will not adopt any language or publication restrictions.
We will identify relevant studies from the Cystic Fibrosis and Genetic Disorders Group's Coagulopathies Trials Register using the term: pregnancy.
The Coagulopathies Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (updated each new issue of The Cochrane Library), quarterly searches of MEDLINE and prospective hand-searching of one specialized journal, Haemophilia. Unpublished work is identified by searching the abstract books of five major conferences: the European Haematology Association conference; the American Society of Hematology conference; the British Society for Haematology Annual Scientific Meeting; the International Society of Haemostasis and Thrombosis Congresses; and the International Congresses of World Federation of Haemophilia. For full details of all searching activities for the register, please see the relevant section of the Cochrane Cystic Fibrosis and Genetic Disorders Group Module.
Trial registries such as ClinicalTrials.gov (http://clinicaltrials.gov/); the International Clinical Trials Registry Platform (ICTRP) (http://www.who.int/ictrp/en/); and the ISRCTN Register (http://www.controlled-trials.com/isrctn) will be searched for ongoing studies.
Searching other resources
We will also check the reference lists of all the studies identified by the above methods.
Data collection and analysis
Selection of studies
Two authors will independently check the titles and abstracts identified from the searches. All authors will obtain the full text of all potentially relevant studies and decide which of these fit the inclusion criteria. We aim to resolve any disagreement by discussion. We will contact study authors for clarification where necessary.
Data extraction and management
Two review authors will independently extract data from included studies using forms provided by the Cochrane Cystic Fibrosis and Genetic Disorders Group. For each included study, we will collect information regarding the patients, the nature of the interventions and data relating to the outcomes specified above. When information regarding any of the above is unclear, we will contact the study authors for further details. We will enter data into the Review Manager software and pool outcome measures where appropriate (at one month, three months, one year and beyond one year) (RevMan 2011).
Assessment of risk of bias in included studies
Two authors (KLK,SK) will independently assess the risk of bias of each included study using the Cochrane Collaboration risk of bias assessment tool (Higgins 2011a). We plan to resolve any disagreements between the two review authors by discussion. The risk of bias of assessment will include: generation of allocation sequence; allocation concealment; blinding (of participants, personnel and outcome assessors); incomplete outcome data, selective outcome reporting; and other potential threats to validity. In relation to each of these domains, we will explicitly judge each of the included studies as having either a low, high or unclear risk of bias (as defined in theCochrane Handbook For Systematic Reviews of Interventions (Higgins 2011a)). For included studies, we will note levels of attrition.
Measures of treatment effect
We will present the results for categorical data as risk ratios (RR) with 95% confidence intervals (CI). For continuous data, if outcomes are measured in the same way, we will use the mean difference (MD) and when the same outcome is measured using different methods, we will use the standardised mean difference (SMD), both with 95% CI.
Unit of analysis issues
We will include cluster-randomised studies in the analyses along with individually randomised studies. In an attempt to account for any unit of analysis error, we plan to use the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b) using an estimate of the intra-cluster correlation co-efficient (ICC) derived from the study (if possible), from a similarly designed study or from a study of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster randomised studies and individually-randomised studies, we plan to pool the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and we consider an interaction between the effect of intervention and the choice of randomisation unit to be unlikely. We will also acknowledge heterogeneity in the randomisation unit and perform a subgroup analysis to investigate the effects of the randomisation unit.
The relevance of cross-over studies in preventing or treating acute haemorrhagic conditions in temporary situations such as pregnancy is minimal as they are usually carried out for stable chronic conditions. In the event that we identify cross-over studies, we will only analyse the data from the first study period.
For multi-arm studies, we will include the intervention groups of relevance in a pair-wise comparison of intervention groups that would meet the criteria for including studies in the review. All intervention groups of a multi-intervention study will be mentioned in the 'Characteristics of included studies' table in either the 'intervention' or 'notes' cell (Higgins 2011a).
Dealing with missing data
For all outcomes, we will carry out analyses, as far as possible, on an intention-to-treat basis, i.e. we will attempt to include all participants randomised to each group in the analyses, and analyse all participants in the group to which they were allocated, regardless of whether or not they received the allocated intervention. Where information is missing or unclear, we will contact the study author(s).
Assessment of heterogeneity
We will assess heterogeneity among studies by inspecting the forest plots and using the Chi
0% to 40%: might not be important;
30% to 60%: may represent moderate heterogeneity;
50% to 90%: may represent substantial heterogeneity;
75% to 100%: considerable heterogeneity.
Final interpretation of the I
Assessment of reporting biases
If any study protocols are available, we will compare these to the published reports. Where the study protocol is not available, we will judge the completeness of reported outcomes based on clinical common sense. If 10 or more studies are eligible, we will investigate potential reporting biases using a funnel plot. We will use a linear regression approach to measure funnel plot asymmetry on the logarithm scale of the RR. If we obtain an asymmetrical funnel plot, we will explore alternative causes in addition to publication bias.
We will perform statistical analysis in accordance with the guidelines developed by the Cochrane Collaboration (Deeks 2011). We will perform our statistical analysis using RevMan 5.1 (RevMan 2011). If there is no significant heterogeneity, we will use the fixed-effect model. In the presence of at least moderate heterogeneity (over 30%), we will use the random-effects model and subgroup analyses as described below to investigate the source of heterogeneity.
Subgroup analysis and investigation of heterogeneity
If we identify substantial heterogeneity, we will investigate it using subgroup analyses as detailed below.
- Congenital and acquired bleeding disorders
- Pre-term and term birth
- Immediate and delayed postpartum haemorrhage
- Type of anaesthesia during c-section (spinal and epidural)
- Types of blood component transfusion in different types of bleeding disorders
We will consider meta-regression when there are more than 10 studies in the meta-analysis.
We will perform a sensitivity analysis (on the primary outcomes only) if we find the included studies are not of 'high quality', i.e. studies defined as having a low risk of bias relating to random sequence generation, adequate allocation concealment and where the percentage of missing data less than 20%, given the stated importance of attrition as a quality measure.
If statistical heterogeneity exists in outcomes, we will carry out a sensitivity analysis to explore the effects of using a fixed- or random-effects analysis. We will also explore the impact of including studies with high levels of missing data in the overall assessment of treatment effect.
Finally, as outlined in the 'Unit of analysis issues' section, if possible, we plan to conduct a sensitivity analysis on the choice of ICC.
We thank the management of Melaka Manipal Medical College, Melaka, Malaysia, Manipal University, India for giving us the opportunity to be involved in the development of this protocol.
The authors would like to thank Tracey Remmington of the Cochrane Cystic Fibrosis and Genetic Disorders Group for the help and support during the course of this protocol.
Contributions of authors
Please complete. NOTE: Order for citation should reflect level of contribution.
Declarations of interest