Aspirin and/or heparin for women with unexplained recurrent miscarriage with or without inherited thrombophilia
Abstract
Background
Since hypercoagulability might result in recurrent miscarriage, anticoagulant agents could potentially increase the chance of live birth in subsequent pregnancies in women with unexplained recurrent miscarriage, with or without inherited thrombophilia.
Objectives
To evaluate the efficacy and safety of anticoagulant agents, such as aspirin and heparin, in women with a history of at least two unexplained miscarriages with or without inherited thrombophilia.
Search methods
We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (1 October 2013) and scanned bibliographies of all located articles for any unidentified articles.
Selection criteria
Randomised and quasi‐randomised controlled trials that assessed the effect of anticoagulant treatment on live birth in women with a history of at least two unexplained miscarriages with or without inherited thrombophilia were eligible. Interventions included aspirin, unfractionated heparin (UFH), and low molecular weight heparin (LMWH) for the prevention of miscarriage. One treatment could be compared with another or with no‐treatment (or placebo).
Data collection and analysis
Two review authors (PJ and SK) assessed the studies for inclusion in the review and extracted the data. If necessary they contacted study authors for more information. We double checked the data.
Main results
Nine studies, including data of 1228 women, were included in the review evaluating the effect of either LMWH (enoxaparin or nadroparin in varying doses) or aspirin or a combination of both, on the chance of live birth in women with recurrent miscarriage, with or without inherited thrombophilia. Studies were heterogeneous with regard to study design and treatment regimen and three studies were considered to be at high risk of bias. Two of these three studies at high risk of bias showed a benefit of one treatment over the other, but in sensitivity analyses (in which studies at high risk of bias were excluded) anticoagulants did not have a beneficial effect on live birth, regardless of which anticoagulant was evaluated (risk ratio (RR) for live birth in women who received aspirin compared to placebo 0.94, (95% confidence interval (CI) 0.80 to 1.11, n = 256), in women who received LMWH compared to aspirin RR 1.08 (95% CI 0.93 to 1.26, n = 239), and in women who received LMWH and aspirin compared to no‐treatment RR 1.01 (95% CI 0.87 to 1.16) n = 322).
Obstetric complications such as preterm delivery, pre‐eclampsia, intrauterine growth restriction and congenital malformations were not significantly affected by any treatment regimen. In included studies, aspirin did not increase the risk of bleeding, but treatment with LWMH and aspirin increased the risk of bleeding significantly in one study. Local skin reactions (pain, itching, swelling) to injection of LMWH were reported in almost 40% of patients in the same study.
Authors' conclusions
There is a limited number of studies on the efficacy and safety of aspirin and heparin in women with a history of at least two unexplained miscarriages with or without inherited thrombophilia. Of the nine reviewed studies quality varied, different treatments were studied and of the studies at low risk of bias only one was placebo‐controlled. No beneficial effect of anticoagulants in studies at low risk of bias was found. Therefore, this review does not support the use of anticoagulants in women with unexplained recurrent miscarriage. The effect of anticoagulants in women with unexplained recurrent miscarriage and inherited thrombophilia needs to be assessed in further randomised controlled trials; at present there is no evidence of a beneficial effect.
PICOs
Plain language summary
Aspirin and/or heparin for women with unexplained recurrent miscarriage with or without inherited thrombophilia
Recurrent miscarriage is associated with inherited blood clotting disorders that could interfere with the placental blood circulation. Recurrent miscarriage can also be unexplained, with no known cause. Anticoagulant drugs such as aspirin or low molecular weight heparin may help women with recurrent miscarriage and such an underlying blood clotting problem. These drugs may also cause bleeding (including nose bleeds and haematomas) in the mother, though not in the baby. Data from nine included randomised controlled trials (involving 1228 women) analysed in this review, provided no evidence to support the use of anticoagulants in women with recurrent miscarriage, regardless of the presence of inherited blood clotting disorders (thrombophilia).
Irrespective of the type or combination of anticoagulant, no benefit of anticoagulant treatment was found for live births. Obstetric complications were not clearly affected by any treatment regimen. Injection of low molecular weight heparin caused local skin reactions (pain, itching, swelling) in one study (side effects were not regularly reported in all studies). In the nine reviewed studies quality varied and different treatments were studied. Three studies were considered at high risk of bias. The number of studies on this topic remains limited.
Thrombophilia refers to blood clotting disorders associated with a predisposition to thrombosis and thus increased risk for thrombotic events. It can be inherited as well as acquired, as is the case in the antiphospholipid syndrome. Both inherited and acquired thrombophilia are associated with vascular thrombosis as well as pregnancy complications including recurrent miscarriage and premature delivery.
Authors' conclusions
Background
Description of the condition
Up to 15% of all clinically recognised pregnancies end in miscarriage (miscarriage before the 20th week of gestational age) (Everett 1997; Huisjes 1984). Approximately 5% of women experience two or more miscarriages (recurrent miscarriage, RM), whereas three or more first trimester miscarriages may affect as many as 1% to 2% of women of reproductive age (Clifford 1994; Cook 1995; Stirrat 1990). RM is devastating for women and their families. The definition of RM remains a subject of debate. The World Health Organization (WHO) defines miscarriage as the spontaneous loss of a clinical pregnancy that occurs before 20 completed weeks of gestational age (WHO 2009). Often RM is defined as three or more consecutive miscarriages. According to recent European Society for Human Reproduction & Embryology (ESHRE) guidelines, RM is traditionally defined as three or more consecutive miscarriages occurring before 20 weeks' gestation (Jauniaux 2006). Recent evidence shows that two miscarriages constitute RM (Bhattacharya 2010; Jaslow 2010). Adequate characterisation of miscarriages and patients in RM studies is most important and, favourably, would make studies mutually comparable (Christiansen 2006). The risk of miscarriage after two or three consecutive miscarriages is similar (Regan 1988).
Furthermore, the presence of parenteral chromosome abnormalities, which is a known risk factor for recurrent miscarriage (Franssen 2005), as well as the presence of antiphospholipid antibodies, another known risk factor for recurrent miscarriage (Jauniaux 2006), are not different in women with two or three miscarriages (van den Boogaard 2010; van den Boogaard 2012). We therefore chose to use the broad definition of RM in this review: two or more not necessarily consecutive miscarriages.
Miscarriage is associated with relevant maternal morbidity like bleeding and infection and, sometimes, maternal death (NHMRC 2001), particularly in low‐income countries (Goyaux 2001). Moreover, miscarriage, especially if recurrent, might cause important psychological and emotional distress that can be further complicated by feelings of anxiety and depression as well as social withdrawal (Lee 1996; Lok 2007).
Several factors may be involved in the aetiology of RM. Women experiencing RM may have an underlying medical condition such as carrier status of a structural chromosome abnormality (Braekeleer 1990; Franssen 2005), antiphospholipid syndrome, or other blood clotting disorders generally referred to as thrombophilias (Robertson 2006) or a septate uterus (Chan 2011). Factors less strongly associated with RM are hyperhomocysteinemia and endocrine abnormalities (Christiansen 2005).
Thrombophilia is a diverse group of coagulation disorders associated with a predisposition to thrombosis and thus increased risk for thrombotic events such as deep vein thrombosis and pulmonary embolism. These hypercoagulable states can either be inherited as the factor V Leiden mutation (which results in a decreased capacity to inactivate activated factor V by the protein C system, also known as activated protein C (APC) resistance), the deficiency of physiological anticoagulants like protein C, protein S and antithrombin and the prothrombin G20210A gene mutation (resulting in increased concentrations of prothrombin in plasma) or an elevated level of factor VIII‐ac (Middeldorp 2007a) or acquired, as for instance the antiphospholipid syndrome. In this latter syndrome, the predisposition to thrombosis is acquired due to the presence of antiphospholipid antibodies (Lim 2006).
A growing body of evidence has implicated thrombophilia in adverse obstetrical events (such as intrauterine growth restriction, (recurrent) miscarriage, severe pre‐eclampsia, and placental abruption) (Kupferminc 1999; Middeldorp 2007), and there is also reasonable evidence to suggest that some cases of RM are associated with thrombosis of placental vessels and infarction. Firstly, microthrombi are a common finding in the placental vasculature of women with RM (Rushton 1988). Secondly, placental thrombosis and infarction have been described in association with certain thrombophilic defects (Dizon 1997; Rai 1996), but other pathophysiological pathways than thrombosis could also be involved, since adverse pregnancy outcomes can occur in women with thrombophilia in the absence of placental thrombosis (Mousa 2000). Thirdly, thrombophilic defects are significantly more prevalent amongst women with such pregnancy complications (Rai 1995; Rey 2003). A meta‐analysis of population‐based studies showed that the magnitude of the association with thrombophilia varies according to the timing of fetal loss (Rey 2003). In particular, first trimester RM was associated with factor V Leiden mutation, APC resistance, and prothrombin G20210A mutation, while late non‐recurrent fetal loss was associated with factor V Leiden mutation, prothrombin G20210A mutation, and protein S deficiency. Also, family studies showed that women with inherited thrombophilia, especially those with combined defects or antithrombin deficiency, have an increased risk of miscarriage and intrauterine fetal death compared to women without these defects (Meinardi 1999; Preston 1996; Sanson 1996).
The antiphospholipid syndrome is an acquired thrombophilia and associated both with vascular thrombosis and pregnancy complications (including RM and premature delivery) (Levine 2002). These adverse pregnancy outcomes may result from placental infarctions and thrombotic changes in decidual microvessels (Infante‐Reviard 1991; Lockshin 1999). The use of antithrombotic agents heparin (unfractionated heparin or low‐molecular‐weight heparin (LMWH)) and aspirin has been studied in women with antiphospholipid syndrome. Both agents have anti‐clotting properties, but these work differently: heparin increases the effect of the natural anticoagulant antithrombin; whereas aspirin inhibits platelet aggregation. Heparin and aspirin seem to be effective and safe in reducing miscarriage rates in women with antiphospholipid syndrome with significantly better pregnancy outcome than aspirin alone (rate of live births of 71% to 80% versus 42% to 44% respectively, an absolute risk difference of 36%) (Kutteh 1996; Rai 1997), although findings have not always been consistent (Farquharson 2002; Laskin 2009). Inconsistency in these study results may be explained by alternate treatment regimens, or the use of different diagnostic criteria, as these have been revised over time by consensus (Miyakis 2006). Both the therapy for RM associated with the antiphospholipid syndrome and other possible therapies currently considered for the prevention of RM (as progesterone and immunotherapy) are the topics covered in other Cochrane reviews (Empson 2005; Haas 2008; Porter 2006).
Estimates of the prognosis of subsequent pregnancies in women with RM without antiphospholipid syndrome range from live birth rates of approximately 50% to 89% (Brigham 1999; Cohn 2010; Lindqvist 2006; Rai 2000; Stirrat 1990). For women with RM and underlying thrombophilic disorders, these figures range from 63% to 80% (Preston 1996; Rai 2000). The differences between studies can probably be explained by differences in the populations of women participating in the studies.
Description of the intervention
The use of anticoagulants in pregnancy needs to be carefully evaluated for efficacy and safety since it can carry risks for the mother and the fetus. Coumarin derivatives are anticoagulant drugs used most often in case of thrombosis, but cross the placenta and can display teratogenic effects. In contrast to coumarin derivatives, neither UFH nor LMWH cross the placenta and therefore do not have the potential to cause fetal bleeding and teratogenicity (Ginsberg 2001). The maternal risks associated with heparin administration are uncommon but potentially serious and include bleeding, heparin‐induced thrombocytopenia and heparin‐induced osteopenia with fractures. Moreover, heparin administration may cause pain and slight bruising at injection sites. There is accumulating evidence that LMWH is at least as effective and safe as UFH with potential advantages during pregnancy, since they cause less heparin‐induced thrombocytopenia, can be administered once daily, and are associated with a lower risk of heparin‐induced osteoporosis (Bates 2012; Ginsberg 2001; Sanson 1999). Based on current evidence, aspirin (less than 150 mg/d) during the second and third trimesters appears to be safe, while the safety of higher doses of aspirin during the first trimester remains uncertain (Askie 2007; Bates 2012; Ginsberg 2001). The use of heparin in pregnancy has been covered in another Cochrane review (Walker 2003).
Why it is important to do this review
In clinical practice, women with RM associated with inherited thrombophilia or RM without any other apparent predisposing disorder are frequently seeking advice about the indication for anticoagulant treatment. Some clinicians tend to extrapolate the beneficial effect of anticoagulant therapy in women with antiphospholipid syndrome and RM to all women with RM. Whether there is evidence for an effect of anticoagulants in women with RM ‐ either unexplained or associated with inherited thrombophilia ‐ is the objective of this review.
Objectives
The objective of this review was to determine whether anticoagulant treatment improves the chance of a live birth in women with a history of at least two unexplained miscarriages with or without inherited thrombophilia.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials and quasi‐randomised controlled trials that assessed the effect of anticoagulant treatment on improving the live birth rate in women with a history of at least two unexplained miscarriages with or without inherited thrombophilia.
Types of participants
Participants were pregnant women with a history of at least two unexplained miscarriages with or without inherited thrombophilia. Of studies that included women attempting to conceive, but who were not pregnant upon randomisation only results of women who became pregnant were included. Studies that included women with apparent risk factors (other than inherited thrombophilia) of RM (antiphospholipid syndrome; uterine abnormalities; patients' or their partners' karyotype abnormalities) were included only if the results from women with a history of at least two unexplained miscarriages with or without inherited thrombophilia could be extracted to be analysed separately. For this review, we selected studies in women with two or more previous miscarriages up to 24 weeks' gestation. The study populations are described whenever possible with regard to number of miscarriages, gestational age of the miscarriages, and maternal age.
Types of interventions
The interventions included were aspirin, UFH, and LMWH for the prevention of miscarriage. One treatment could be compared with another or with no treatment (or placebo). Combinations of therapy could be used. To exclude a potential lack of efficacy due to a limited duration of treatment, only studies in which the investigational treatment was started at a maximum of 12 weeks' gestation and continued beyond 32 weeks' gestation or until the end of pregnancy were eligible.
Types of outcome measures
Primary outcomes
Live birth.
Secondary outcomes
-
Preterm delivery of a live infant before 37 weeks' gestational age (not a prespecified outcome).
-
Preterm delivery of a live infant between 24 and 28 weeks' gestational age.
-
Preterm delivery of a live infant between 28 and 32 weeks' gestational age.
-
Preterm delivery of a live infant between 32 and 37 weeks' gestational age.
-
Obstetric complications (gestational hypertension, pre‐eclampsia, intrauterine growth restriction).
-
Congenital malformations.
-
Admission to special care.
-
Side effects of the drug used, both for the mother and the baby (maternal and/or neonatal bleeding, heparin‐induced thrombocytopenia, heparin‐induced osteopenia, pain, itching or swelling at injection sites and allergic reactions to heparin).
-
Thromboembolic complications.
The following methods section of this review is based on a standard template used by the Cochrane Pregnancy and Childbirth Group.
Search methods for identification of studies
Electronic searches
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co‐ordinator (1 October 2013).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co‐ordinator and contains trials identified from:
-
monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
-
weekly searches of MEDLINE;
-
weekly searches of Embase;
-
handsearches of 30 journals and the proceedings of major conferences;
-
weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co‐ordinator searches the register for each review using the topic list rather than keywords.
For additional searching performed for the previous version of the review, please see Appendix 1.
Searching other resources
We scanned bibliographies of all located articles for further studies.
We did not apply any language restrictions.
Data collection and analysis
For the methods used when assessing the trials identified in the previous version of this review, seeAppendix 2.
For this update we used the following methods when assessing the reports identified by the updated search.
Selection of studies
Two review authors independently assessed for inclusion all the potential studies we identified as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted a third review author.
Data extraction and management
We designed a form to extract data. For eligible studies, two review authors extracted the data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted a third review author. We entered data into Review Manager software (RevMan 2012) and checked for accuracy.
When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.
Dr Kaandorp, Dr Goddijn, and Dr Middeldorp were investigators of the randomised controlled trial ALIFE study (Kaandorp 2010) and so this trial was assessed by the other review authors.
Assessment of risk of bias in included studies
Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third assessor.
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
-
low risk of bias (any truly random process, e.g. random number table; computer random number generator);
-
high risk of bias (any non‐random process, e.g. odd or even date of birth; hospital or clinic record number);
-
unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal allocation to interventions prior to assignment and assessed whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
-
low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
-
high risk of bias (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth);
-
unclear risk of bias.
(3.1) Blinding of participants and personnel (checking for possible performance bias)
Performance bias and detection bias were not included as criterion for quality as we anticipated that the outcome live birth was not influenced by blinding.
(3.2) Blinding of outcome assessment (checking for possible detection bias)
Performance bias and detection bias were not included as criterion for quality as we anticipated that the outcome live birth was not influenced by blinding.
(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We state whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we re‐included missing data in the analyses which we undertook.
We assessed methods as:
-
low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups; ≦ 20% participants missing);
-
high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation; more than 20% participants missing);
-
unclear risk of bias.
(5) Selective reporting (checking for reporting bias)
We described for each included study how we investigated the possibility of selective outcome reporting bias.
We assessed the methods as:
-
low risk of bias (where it is clear that all of the study’s pre‐specified outcomes and all expected outcomes of interest to the review have been reported);
-
high risk of bias (where not all the study’s pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);
-
unclear risk of bias.
(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)
We described for each included study any important concerns we have about other possible sources of bias.
We assessed whether each study was free of other problems that could put it at risk of bias:
-
low risk of other bias;
-
high risk of other bias;
-
unclear whether there is risk of other bias.
(7) Overall risk of bias
We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it likely to impact on the findings. We explored the impact of the level of bias through undertaking sensitivity analyses ‐ see Sensitivity analysis.
Measures of treatment effect
Dichotomous data
For dichotomous data, we presented results as summary risk ratio with 95% confidence intervals.
Continuous data
We did not analyse any continuous data.
Unit of analysis issues
Cluster‐randomised trials
We did not identify any cluster‐randomised trials for inclusion in this review. In future updates, if identified and eligible, we will include cluster‐randomised trials in the analyses along with individually‐randomised trials. We will adjust their sample sizes using the methods described in the Handbook [Section 16.3.4 or 16.3.6] using an estimate of the intracluster correlation co‐efficient (ICC) derived from the trial (if possible), from a similar trial 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 trials and individually‐randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely. We will also acknowledge heterogeneity in the randomisation unit and perform a sensitivity analysis to investigate the effects of the randomisation unit.
Cross‐over trials
Cross‐over trials were excluded.
Dealing with missing data
For included studies, we noted levels of attrition. We planned to explore the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis. However, none of the included studies were considered to be at high risk of attrition bias.
For all outcomes, we carried out analyses, as far as possible, on an intention‐to‐treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta‐analysis using the T², I² and Chi² statistics. We regarded heterogeneity as substantial if an I² was greater than 30% and either a T² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.
Assessment of reporting biases
In future updates, if there are 10 or more studies in the meta‐analysis we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.
Data synthesis
Data were pooled based on type of intervention, but irrespective of the dose of LMWH or aspirin. Only studies in which the investigational treatment was started at a maximum of 12 weeks' gestation and continued beyond 32 weeks' gestation were included.
We carried out statistical analysis using the Review Manager software (RevMan 2012). We used fixed‐effect meta‐analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we used random‐effects meta‐analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random‐effects summary was treated as the average of the range of possible treatment effects and we planned to discuss the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we did not combine trials.
If we used random‐effects analyses, the results were presented as the average treatment effect with 95% confidence intervals, and the estimates of Tau² and I².
Subgroup analysis and investigation of heterogeneity
If we identified substantial heterogeneity, we planned to investigate it using subgroup analyses and sensitivity analyses. We planned to consider whether an overall summary was meaningful, and if it was, use random‐effects analysis to produce it.
We planned to carry out the following subgroup analyses:
-
inherited thrombophilia versus no inherited thrombophilia;
-
different inherited thrombophilic disorders;
-
preconceptional versus periconceptional anticoagulant use;
-
type of anticoagulant(s) used (e.g. single drug, combination of anticoagulant agents);
-
dose of anticoagulant(s);
-
duration of anticoagulant use;
-
women with a history of three or more miscarriages;
-
women with no previous live birth versus women with one or more previous live birth.
We planned to use the primary outcome, live birth, in subgroup analyses. However, due to lack of data, we were not able to conduct planned subgroup analysis. We included data from single subgroups in the analysis of the primary outcome, but due to lack of data, we did not explore the treatment effect between the pre‐specified subgroups as outlined above.
In future updates, if more data become available, we will assess subgroup differences by interaction tests available within RevMan (RevMan 2012) and report the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.
Sensitivity analysis
We performed sensitivity analyses for the main outcomes by individual quality criteria to assess the effect of poorer quality studies on the magnitude of the estimate of effect. Only studies at an overall low risk of bias were included in the initial analyses and we carried out sensitivity analyses to explore the effect of quality.
Results
Description of studies
Included studies
Details for the studies included are in the Characteristics of included studies.
We included nine studies (1228 women for the primary outcome live birth) (Badawy 2008; Clark 2010; Dolitzky 2006; Fawzy 2008; Giancotti 2012Kaandorp 2010; Martinelli 2012; Tulppala 1997; Visser 2011) in this review. In the studies Badawy 2008 and Dolitzky 2006, full study populations were included. For the studies Clark 2010; Fawzy 2008; Giancotti 2012; Kaandorp 2010; Martinelli 2012; Tulppala 1997; and Visser 2011, we could not include full study populations, but had to extract data on the women fulfilling the inclusion criteria of the review. Reasons for including only part of the original study population are explained for each study.
Badawy 2008 evaluated the effect of LMWH (enoxaparin 20 mg/day) in women with three or more consecutive first trimester miscarriages. Women were included if no risk factors for RM could be identified (women with inherited thrombophilia were excluded) and randomised to either treatment or no‐treatment (no placebo). Therapy was commenced once fetal viability was detected on ultrasound and continued until 34 weeks' gestation. The primary outcome was pregnancy loss and pregnancy complications, but live births could be calculated from the report and were confirmed by the study author. Of 350 women enrolled, 10 women (four (2.3%) and six (3.4%) in both arms) were lost to follow‐up, leaving 170 women in each study arm for analysis. Side effects of treatment were only reported for women randomised to treatment, not in those randomised to no‐treatment.
Clark 2010 studied the effect of LMWH (enoxaparin 40 mg/day) and aspirin (75 mg/day) from before seven weeks' gestation until 36 weeks' gestation on live birth in 296 women with RM, defined as a minimum of two consecutive early pregnancy losses (at or before 24 weeks' gestation). Intervention with LMWH and aspirin combined with intense pregnancy surveillance was compared with intense pregnancy surveillance without pharmacological intervention. Women were included upon a positive pregnancy test (before seven weeks' gestation). Investigations for uterine or chromosomal abnormalities and antiphospholipid syndrome were conducted only for women with three or more previous miscarriages. Women who were included because of two previous miscarriages were excluded from our analysis, because it could not be confirmed that previous miscarriages were 'unexplained'. The primary outcome measure in the study was live birth. Adverse events of intervention were reported, though could not be extracted for the women with three or more previous miscarriages. One‐hundred and twenty‐two women were included in the review (64 randomised to LMWH and aspirin versus 58 randomised to surveillance).
Dolitzky 2006 evaluated the effect of LMWH compared with aspirin in 104 women with unexplained RM. RM was defined as three or more consecutive first trimester miscarriages or at least two consecutive second trimester miscarriages. The objective was to compare the effect of enoxaparin and aspirin on live birth rate. Women were only included if there was no apparent risk factors for the miscarriages and women with inherited thrombophilia were excluded. The treatment with enoxaparin (40 mg/day) or aspirin (100 mg/day) was started from the time of detection of a fetal heart beat at six to 12 weeks' gestation and continued until a gestational age of 37 weeks. Of the 107 included women, 54 received enoxaparin, 50 aspirin and three were lost to follow‐up. Besides the primary outcome measure of live birth, secondary outcomes like preterm delivery, intrauterine growth restriction, and pre‐eclampsia were reported.
Fawzy 2008 assessed the effect of LMWH (enoxaparin 20 mg/day) compared with combination treatment (oral prednisone and progesterone from the onset of pregnancy until 12 weeks of gestation and aspirin from the onset of pregnancy until 32 weeks of gestation) compared with placebo (for oral intervention) in women with three or more consecutive unexplained miscarriages (before 24 weeks' gestation) with the same partner. Women with inherited thrombophilia were excluded. From this study, we extracted data for the 107 women receiving enoxaparin or placebo. Of these women, 57 were assigned to enoxaparin and 50 to placebo. Treatment was started when a fetal pole was detected and continued until term. The primary outcome was live birth, but secondary outcomes such as obstetric complications and neonatal outcomes were also reported.
Giancotti 2012 evaluated the effect of LMWH or aspirin or a combination of these in 167 women with a history of two or more unexplained miscarriages before 12 weeks' gestation. Women with uterine or chromosomal abnormalities were excluded, but women with inherited thrombophilia or antiphospholipid syndrome were eligible. Women were randomised to LMWH (enoxaparin 40 mg/day from diagnosis of intrauterine pregnancy until delivery) or aspirin (100 mg/day from diagnosis of pregnancy until 32 weeks' gestation) or first aspirin (100 mg/day from diagnosis of pregnancy), which was replaced by LMWH at 32 weeks' gestation (enoxaparin 40 mg/day continued until delivery). For this analysis, we included only data from women without antiphospholipid antibodies randomised to either LMWH (n = 40) or aspirin (n = 46). The primary outcome of the study was live birth, and no secondary outcome measures were reported.
Kaandorp 2010 evaluated the effect of open label LMWH (nadroparin 2850 IU/day) combined with aspirin (80 mg/day) or aspirin only (80 mg/day) compared with placebo in 364 women with unexplained RM with or without inherited thrombophilia. Previous miscarriage was defined as pregnancy loss at a gestational age of 20 weeks or less. Women were included in the study if they were attempting to conceive or were less than six weeks pregnant. From this study, we extracted data of the 299 women who became pregnant (97 were assigned to aspirin plus nadroparin, 99 were assigned to aspirin only and 103 were assigned to placebo). LMWH was initiated when a viable intrauterine pregnancy was confirmed on ultrasonography at six weeks' gestation until the start of labour. Aspirin or placebo was started at randomisation and continued until a gestational age of 36 weeks. The primary outcome was live birth, and secondary outcomes were adverse pregnancy outcomes and maternal adverse events. Secondary outcomes such as obstetric complications and neonatal events were evaluated for 200 women with ongoing pregnancy beyond 12 weeks of gestation.
Martinelli 2012 evaluated the effect of open label LMWH (nadroparin 3800 IU/day), compared with no treatment in 135 women with previous placenta‐mediated pregnancy complications. Women with antiphospholipid syndrome, uterine anomalies or abnormal karyotype were excluded from the study. Inherited thrombophilia was not an exclusion criterion. LMWH was compared with medical surveillance only and treatment was initiated upon randomisation and continued until delivery. Randomisation was performed around the 12th week of gestation, after pregnancy was confirmed on ultrasonography. The primary outcome of the study was a composite outcome of several pregnancy complications. For this review we included data of six women, who had a history of two or more unexplained miscarriages up to 24 weeks' gestation.
Tulppala 1997 evaluated the effect of aspirin (50 mg/day) on live birth rate in 66 pregnant women with preceding RM with or without detectable anticardiolipin antibodies and no other apparent risk factors for their previous miscarriages. RM was defined as three or more consecutive miscarriages (occurring before 22 weeks of gestational age). Aspirin was compared with placebo, and medication was started as soon as a home urinary pregnancy test became positive and continued until delivery. From this study, we extracted data for 54 women who were negative for anticardiolipin antibodies. Of these, 27 were assigned to aspirin and 27 to placebo. Secondary outcomes, such as preterm delivery, obstetric complications, and bleeding rate could not be extracted separately for the group of women with negative anticardiolipin antibodies.
Visser 2011 evaluated the effect of LMWH (enoxaparin 40 mg/day) plus oral placebo (n = 68) compared with LMWH (enoxaparin 40 mg/day) plus aspirin (100 mg/day) (n = 63), compared with aspirin only (100 mg/day) (n = 76) in women with unexplained RM with or without inherited thrombophilia. RM was defined as three or more consecutive first trimester miscarriages, two or more second trimester miscarriages or one third trimester fetal loss combined with at least one first trimester miscarriage. Treatment was initiated upon randomisation (before seven weeks' gestation); aspirin and placebo were discontinued at 36 weeks' gestation whereas enoxaparin was continued until the first signs of labour. The primary outcome was live birth and secondary outcomes were adverse pregnancy outcomes and bleeding. Premature delivery, obstetrical complications and congenital malformations were reported only for women who had live birth. Study authors stated that only one woman was included in the study because of RM based on one fetal loss before, and two fetal losses after 24 weeks' gestation; her data were excluded from this review. Furthermore, data of 10 women (allocated to enoxaparin plus placebo (n = 3, one live birth), enoxaparin plus aspirin (n = 2, one live birth), aspirin only (n = 5, four live births)) were excluded because of the presence of Beta‐2 glycoprotein antibodies. One‐hundred and ninety‐six women were included in this trial.
As can be noted, no study compared the same treatment regimen. Studies compared different doses of LMWH and aspirin or combinations of these, and treatment was started at various gestational ages. As described in the Methods section, we pooled data based on type of intervention, but irrespective of the dose of LMWH or aspirin. Only studies in which the investigational treatment was started at a maximum of 12 weeks' gestation and continued beyond 32 weeks' gestation were included.
Excluded studies
Overall, we excluded 22 studies from the review. We have provided the reasons for exclusion in the Characteristics of excluded studies table.
Three studies are only available in abstract form (Rodger 2009; Salman 2012; Schleussner 2013) and we are awaiting the full study report, see Characteristics of studies awaiting classification.
Risk of bias in included studies
Details for the included studies are shown in the Characteristics of included studies and in Figure 1.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
The studies by Clark 2010; Kaandorp 2010; Martinelli 2012 and Visser 2011 were considered to be at low risk of bias for all assessed criteria.
Allocation
Procedures for adequate allocation concealment were well described in the studies by Badawy 2008; Clark 2010; Dolitzky 2006; Giancotti 2012; Kaandorp 2010; Martinelli 2012; and Visser 2011. Dr Tulppala provided information about allocation in her study (Tulppala 1997), which was then judged to be at low risk of bias. Based on the report and information provided by Dr Fawzy regarding treatment allocation, we judged this study (Fawzy 2008) to be at high risk of selection bias.
Blinding
As explained in the methods sections, performance bias and detection bias were not included as criterion for quality.
Incomplete outcome data
The study by Kaandorp 2010 was analysed on intention‐to‐treat basis, including women lost to follow‐up. Other studies reported outcomes for randomised women minus any participants whose outcomes were missing. Numbers of women lost to follow‐up were small and well balanced for each group and not enough to have a clinically relevant impact on the intervention effect estimate. All studies were therefore considered to be at low risk of attrition bias.
Selective reporting
Assessment of reporting bias was impossible for Dolitzky 2006; and Tulppala 1997 as trial registration was not operative at time of inclusion.The studies by Badawy 2008, Fawzy 2008 and Giancotti 2012 were considered to be at high risk of bias because they were not registered in a prospective trial register. Other studies were considered to be at low risk of reporting bias.
Other potential sources of bias
There were inconsistencies in the report by Badawy 2008, and the report by Giancotti 2012 provided no baseline table, which made it impossible to determine whether prognostic factors were evenly distributed between groups.
Effects of interventions
We included nine studies, involving 1228 participants, in this review. Since treatment regimens varied among included studies, pooled analysis could not include more than three studies, except for when LMWH with or without aspirin was compared to no treatment. Only one study (Kaandorp 2010) included women who used anticoagulants preconceptionally. This yielded insufficient data to perform the planned subgroup analysis for pre‐ and periconceptional anticoagulant use. Where studies reported pregnancy complications, different denominators (e.g. all pregnant women, only ongoing pregnancies, only women with live births) were used in different studies and results could not be pooled. This is explained for the comparisons of treatment concerned.
Aspirin versus no treatment
Pooled analysis from 256 patients showed that compared to placebo, aspirin did not increase live birth (risk ratio (RR) 0.94, 95% confidence interval (CI) 0.80 to 1.11), Analysis 1.1, (Kaandorp 2010; Tulppala 1997). Subgroup analyses for the outcome live birth for women with no previous live births (RR 0.93, 95% CI 0.72 to 1.21), inherited thrombophilia (RR 1.08, 95% CI 0.63 to 1.85) or more than two miscarriages (RR 0.95, 95% CI 0.70 to 1.28) could only be performed for the study by Kaandorp 2010 and also showed no effect of treatment.
Secondary outcomes were not reported by Tulppala 1997. Preterm delivery, obstetric complications and congenital malformations as reported by Kaandorp 2010 for women with ongoing pregnancies beyond 12 weeks' gestation did not differ between the two groups. Bleeding as a side effect from treatment (mainly nose or gum bleeds or haematomas) was reported for 30% of women receiving aspirin and for 27% women receiving placebo (RR 1.11, 95% CI 0.72 to 1.72, Analysis 1.9). It should be noted that bleeding was reported for all women included in the study, including women who did not become pregnant during the course of the study.
LMWH versus aspirin
Three studies compared enoxaparin with aspirin (Dolitzky 2006; Giancotti 2012; Visser 2011). Pooled analysis (n = 325) showed an average RR of live birth for women treated with aspirin of 1.16 (95% CI 0.93 to 1.45; Heterogeneity: Tau² = 0.03; I² = 67%, Analysis 3.1). Due to substantial statistical heterogeneity being detected, we used random‐effects meta‐analysis in Analysis 3.1. After excluding the study by Giancotti 2012 at high risk of bias, both groups had similar live birth rates, 76% in the enoxaparin group and 70% in the aspirin group (RR 1.08, 95% CI 0.93 to 1.26, Analysis 2.1). In the subgroup of women with no previous live births, the RR of live birth with LMWH was 1.24 (95% CI 1.02 to 1.49) compared to aspirin.
The study by Giancotti 2012 reported no secondary outcome measures. Results of secondary outcome measures for the studies by Dolitzky 2006 and Visser 2011 could not be pooled, as both studies used different denominators (i.e. all pregnancies in the study by Dolitzky 2006 and only women with live birth in the study by Visser 2011). In the individual studies, the number of preterm deliveries (before 37 weeks), cases of intrauterine growth restriction, pre‐eclampsia and congenital malformations did not differ between the two groups, Bleeding complications did not differ between the two groups in both studies, though results of bleeding were very different; 0% versus 0.04% in the study by Dolitzky 2006 and 49% versus 50% in the study by Visser 2011 in women treated with LMWH versus aspirin respectively.
LMWH versus no treatment
The effect of LMWH was evaluated in three studies (Badawy 2008; Fawzy 2008; Martinelli 2012). Pooled analysis (n = 453) showed an average RR of live birth for women treated with LMWH of 1.23 (95% CI 0.84 to 1.81; Heterogeneity: Tau² = 0.09; I² = 80%, Analysis 5.1). Due to substantial statistical heterogeneity being detected, we used random‐effects meta‐analysis in Analysis 5.1. When excluding studies at high risk of bias, only the data of six women included in the study by Martinelli 2012 could be analysed. Of these six women, four were randomised to LMWH and two to no treatment and all six had a live birth during the study (Analysis 4.1). Secondary outcomes were only reported by studies at high risk of bias, and could not be pooled due to a difference in denominators (pregnancies continued beyond 21 weeks in the study by Badawy 2008 and all pregnancies in the study by Fawzy 2008). No difference between treatment groups were found in individual studies for pregnancy complications, bleeding or thromboembolic events. The study by Badawy 2008 reported that 22% of women treated with LMWH experienced symptoms of bleeding and 30% local skin reactions.
LMWH and aspirin versus no treatment
The effect of LMWH and aspirin on live birth compared to no treatment or placebo was evaluated in two studies (n = 322) (Clark 2010; Kaandorp 2010). Live birth occurred as often in women receiving LMWH and aspirin (n = 161) as in women who received no treatment (n = 161) (RR for women who received LMWH and aspirin 1.01, 95% CI 0.87 to 1.16), Analysis 6.1). Subgroup analyses for the outcome live birth could only be performed for the study by Kaandorp 2010. For women with no previous live births (RR 1.05, 95% CI 0.83 to 1.34), or more than two miscarriages ( RR 1.00, 95% CI 0.75 to 1.33) no effect of treatment was found. Data of women with inherited thrombophilia suggested a potential benefit in these women when treated with LMWH and aspirin, but the subgroup was underpowered for firm conclusions (RR 1.25, 95% CI 0.74 to 2.12).
Data of secondary outcomes were not available for the study by Clark 2010 and are therefore only described for the study by Kaandorp 2010. The occurrence of obstetric complications did not differ between the two study arms. Maternal bleeding (mainly nose or gum bleeds or haematomas) occurred significantly more frequently in women who received treatment; resulting in a RR of any bleeding of 2.28 (95% CI 1.60 to 3.24, Analysis 6.9). Pain, itching and swelling at injection site was reported by 51% of the women treated with LMWH and aspirin.
LMWH and aspirin versus aspirin
In the studies by Kaandorp 2010 and Visser 2011 the effect of LMWH and aspirin was compared with aspirin only (n = 327). Live birth did not differ significantly between both groups; 68% and 61% respectively (RR 1.11, 95% CI 0.94 to 1.30, Analysis 7.1). Subgroup analyses for the outcome live birth in women with either no previous live births, inherited thrombophilia or more than two previous miscarriages (study by Kaandorp 2010 only) also showed no effect of treatment.
Again, results for secondary outcome measures could not be pooled because different denominators (ongoing pregnancies in the study by Kaandorp 2010 and pregnancies that ended in live birth in the study by Visser 2011) were used. The incidence of preterm delivery, pre‐eclampsia, intrauterine growth restriction (IUGR) and congenital malformations was similar in both groups in the individual studies. A significant difference was seen in bleeding (mainly nose or gum bleeds or haematomas) between the two groups, favouring treatment with aspirin only (RR for bleeding in women treated with LMWH and aspirin 2.04, 95% CI 1.46 to 2.86, Analysis 7.9), in the study by Kaandorp 2010, 62% of women treated with LMWH and aspirin experienced bleeding compared to 30% in women treated with aspirin only. In the study by Visser 2011, bleeding (reported as first trimester, second/third trimester bleeding or postpartum haemorrhage) was reported more often in women treated with aspirin (38% in women treated with LMWH and aspirin and 50% in women treated with aspirin only), though this difference was not significant (RR 0.75, 95% CI 0.45 to 1.24, Analysis 7.9).
LMWH with aspirin versus LMWH
One study evaluated the effect of LMWH and aspirin (n = 61) in comparison with LMWH only (n = 65) (Visser 2011). Neither live birth, nor any of the secondary outcomes including bleeding were different between the two groups (RR of live birth in women treated with LMWH plus aspirin 0.91, 95% CI 0.72 to 1.15, Analysis 8.1). Subgroup analyses in women with inherited thrombophilia or no previous live birth were small and showed no benefit of one treatment over the other.
LMWH with or without aspirin versus no treatment
Results of studies were combined, to evaluate the effect of LMWH with or without aspirin on the chance of live birth. Pooled results from 793 patients of five studies (Badawy 2008; Clark 2010; Fawzy 2008; Kaandorp 2010; Martinelli 2012) showed no effect of treatment (RR 1.07, 95% CI 0.99 to 1.15, Analysis 9.1). After excluding studies at high risk of bias the point estimate shifted towards 1 and no effect of treatment was observed (n = 324, RR for live birth in women treated with LMWH 0.98, 95% CI 0.85 to 1.12, Analysis 10.1).
Discussion
Since the last update of this review (Kaandorp 2009), seven randomised controlled trials on the efficacy and safety of aspirin and heparin in women with a history of at least two miscarriages without apparent risk factors other than inherited thrombophilia were published, but the number of studies on this topic remains limited. Although in total nine studies were included, three studies were considered to be at high risk of bias and from one study, data of only six women could be included. Different treatment doses of anticoagulants compared, prescribed for different time periods, resulted in maximally three studies per comparison that could be pooled and only three of seven comparisons included a placebo‐ or no treatment‐arm.
Irrespective of the type of or combinations of anticoagulants used, no benefit of anticoagulant treatment for live birth was found.
Data for subgroup analyses were scarce. Subgroup analyses in women with more than two previous miscarriages showed no effect of treatment, regardless which treatment regimens were compared. A trend towards a significant effect from LMWH when compared to aspirin (risk ratio (RR) of live birth 1.21, 95% confidence interval (CI) 0.79 to 1.87) and of LMWH and aspirin when compared to no treatment (RR of live birth 1.25, 95% CI 0.74 to 2.12) was observed in women with inherited thrombophilia but the subgroups were underpowered for firm conclusions. As the clinical question of efficacy of anticoagulants for women with recurrent miscarriage (RM) and inherited thrombophilia remains relevant, randomised controlled trials focussing on women with inherited thrombophilia only are urgently needed.
In subgroup analyses of women with no previous live birth, a beneficial effect of LMWH over aspirin was found in pooled analyses of two studies (n = 112, RR 1.24, 95% CI 1.02 to 1.49). Some evidence of a similar trend toward a beneficial effect for LMWH versus LMWH and aspirin was observed in a small subgroup in one study (n = 72, RR of live birth in women treated with LMWH and aspirin 0.77, 95% CI 0.59 to 1.02). In comparisons of LMWH and aspirin with either no treatment or with aspirin, no beneficial effect of LMWH and aspirin was found for women with no previous live birth. All subgroup analyses in women with no previous live birth were limited due to small numbers.
Reporting of secondary outcomes varied widely among studies. In studies that reported pregnancy complications, different denominators (e.g. all pregnant women, only ongoing pregnancies, only live births) were used and results could not be pooled. In the individual studies, no effect of treatment on any pregnancy complication was found. Especially reporting of side effects of treatment, such as bleeding and local skin reactions (for LMWH) was inconsistent. The risk of bleeding (mainly nose or gum bleeds or haematomas) was more than two‐fold increased in women treated with LMWH and aspirin, compared to either aspirin only or no treatment in the study by Kaandorp 2010. This was not confirmed in the study by Visser 2011 (reporting first‐trimester, second/third trimester bleeding and postpartum haemorrhage, results could not be pooled). Local skin reactions to administered LMWH were not reported in several studies, whereas 50% of women treated with LMWH and aspirin in the study by Kaandorp 2010 reported these side effects.
Besides the limited number of studies on this topic, we must appreciate the heterogeneity between studies. Study designs vary regarding dose, duration and type of treatment, blinding, reporting of secondary outcomes and study quality. Furthermore, lack of a no‐treatment (or placebo) arm in some studies impedes assessing a risk‐benefit ratio for the individual interventions.
Although in approximately half of women with RM risk factors can be identified, in 50% this remains unexplained. One could argue that the group of women with unexplained RM is in fact a heterogeneous collection of subgroups of women with as yet unidentified risk factors. Once identified, these risk factors may render them ineligible for the study. For this review, we have studied women with unexplained RM (with or without inherited thrombophilia) as one group and where possible we performed a subgroup analysis in women with inherited thrombophilia. Obvious risk factors of RM such as antiphospholipid syndrome, abnormal parental karyotypes and uterine abnormalities were exclusion criteria. This approach yields results that can be applied to all women with unexplained RM, with or without inherited thrombophilia. If future research leads to the identification of other risk factors for RM, therapeutic options for that newly identified subgroup will have to be considered.
Our review of currently available (though limited) data, does not support the use of anticoagulants heparin, LMWH and aspirin, or combinations of these, for women with unexplained RM with or without inherited thrombophilia. Included studies were not sufficiently powered to evaluate an effect of heparin, LMWH or aspirin or combinations in women with confirmed inherited thrombophilia and unexplained recurrent miscarriage. Large randomised controlled trials assessing this subgroup are urgently needed, but until results are available, anticoagulants in these women are also not recommended in this setting.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Aspirin versus no treatment, Outcome 1 Live birth.
Comparison 1 Aspirin versus no treatment, Outcome 2 Preterm delivery < 37 weeks.
Comparison 1 Aspirin versus no treatment, Outcome 3 Preterm delivery 24‐28 weeks.
Comparison 1 Aspirin versus no treatment, Outcome 4 Preterm delivery 28‐32 weeks.
Comparison 1 Aspirin versus no treatment, Outcome 5 Preterm delivery 32‐37 weeks.
Comparison 1 Aspirin versus no treatment, Outcome 6 Obstetric complications; pre‐eclampsia.
Comparison 1 Aspirin versus no treatment, Outcome 7 Obstetric complications; IUGR.
Comparison 1 Aspirin versus no treatment, Outcome 8 Congenital malformations.
Comparison 1 Aspirin versus no treatment, Outcome 9 Side effects; any bleeding.
Comparison 1 Aspirin versus no treatment, Outcome 10 HIT/thrombocytopenia.
Comparison 2 LMWH versus aspirin, Outcome 1 Live birth.
Comparison 2 LMWH versus aspirin, Outcome 2 Preterm delivery < 37 weeks.
Comparison 2 LMWH versus aspirin, Outcome 3 Preterm delivery 24‐28 weeks.
Comparison 2 LMWH versus aspirin, Outcome 4 Preterm delivery 28‐32 weeks.
Comparison 2 LMWH versus aspirin, Outcome 5 Preterm delivery 32‐37 weeks.
Comparison 2 LMWH versus aspirin, Outcome 6 Obstetric complications; pre‐eclampsia.
Comparison 2 LMWH versus aspirin, Outcome 7 Obstetric complications; IUGR.
Comparison 2 LMWH versus aspirin, Outcome 8 Obstetric complications; congenital malformations.
Comparison 2 LMWH versus aspirin, Outcome 9 Side effects; any bleeding.
Comparison 3 LMWH versus aspirin including studies at high risk of bias, Outcome 1 Live birth.
Comparison 4 LMWH versus no treatment, Outcome 1 Live birth.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 1 Live birth.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 2 Preterm delivery < 37 weeks.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 3 Obstetric complications; pre‐eclampsia.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 4 Obstetric complications; IUGR.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 5 Obstetric complications; congenital malformations.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 6 Side effects; any bleeding.
Comparison 5 LMWH versus no treatment including studies at high risk of bias, Outcome 7 Thromboembolic complications.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 1 Live birth.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 2 Preterm delivery < 37 weeks.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 3 Preterm delivery 24‐28 weeks.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 4 Preterm delivery 28‐32 weeks.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 5 Preterm delivery 32‐37 weeks.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 6 Obstetric complications; pre‐eclampsia.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 7 Obstetric complications; IUGR.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 8 Congenital Malformations.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 9 Side effects; any bleeding.
Comparison 6 LMWH and aspirin versus no treatment, Outcome 10 Side effects; pain/itching/swelling at injection site.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 1 Live birth.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 2 Preterm delivery < 37 weeks.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 3 Preterm delivery 24‐28 weeks.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 4 Preterm delivery; 28‐32 weeks.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 5 Preterm delivery; 32‐37 weeks.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 6 Obstetric complications; pre‐eclampsia.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 7 Obstetric complications; IUGR.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 8 Congenital malformations.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 9 Side effects; any bleeding.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 10 Side effects; pain/itching/swelling at injection site.
Comparison 7 LMWH and aspirin versus aspirin, Outcome 11 HIT/thrombocytopenia.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 1 Live birth.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 2 Preterm delivery < 37 weeks.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 3 Obstetric complications; pre‐eclampsia.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 4 Obstetric complications; IUGR.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 5 Congenital malformations.
Comparison 8 LMWH and aspirin versus LMWH, Outcome 6 Side effects; any bleeding.
Comparison 9 LMWH with or without aspirin vs no treatment including studies at high risk of bias, Outcome 1 Live birth.
Comparison 10 LMWH with or without aspirin vs no treatment, Outcome 1 Live birth.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Live birth in all women | 2 | 256 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.94 [0.80, 1.11] |
| 1.2 Live birth in women with no previous live birth | 1 | 122 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.93 [0.72, 1.21] |
| 1.3 Live birth in women with inherited thrombophilia | 1 | 32 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.08 [0.63, 1.85] |
| 1.4 Live birth in women with more than two previous miscarriages | 1 | 117 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.95 [0.70, 1.28] |
| 2 Preterm delivery < 37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Preterm delivery 24‐28 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Preterm delivery 28‐32 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Preterm delivery 32‐37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Obstetric complications; pre‐eclampsia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Obstetric complications; IUGR Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 8 Congenital malformations Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 9 Side effects; any bleeding Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 10 HIT/thrombocytopenia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Live birth in all women | 2 | 239 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.08 [0.93, 1.26] |
| 1.2 Live birth in women with no previous live birth | 2 | 112 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.24 [1.02, 1.49] |
| 1.3 Live birth in women with inherited thrombophilia | 1 | 36 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.21 [0.79, 1.87] |
| 2 Preterm delivery < 37 weeks Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Preterm delivery 24‐28 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Preterm delivery 28‐32 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Preterm delivery 32‐37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Obstetric complications; pre‐eclampsia Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Obstetric complications; IUGR Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 8 Obstetric complications; congenital malformations Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 9 Side effects; any bleeding Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 3 | 325 | Risk Ratio (M‐H, Random, 95% CI) | 1.16 [0.93, 1.45] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 1 | 6 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.0 [0.56, 1.79] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 3 | 453 | Risk Ratio (M‐H, Random, 95% CI) | 1.23 [0.84, 1.81] |
| 2 Preterm delivery < 37 weeks Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Obstetric complications; pre‐eclampsia Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Obstetric complications; IUGR Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Obstetric complications; congenital malformations Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Side effects; any bleeding Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Thromboembolic complications Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Live birth in all women | 2 | 322 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.87, 1.16] |
| 1.2 Live birth in women with no previous live birth | 1 | 118 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.05 [0.83, 1.34] |
| 1.3 Live birth in women with inherited thrombophilia | 1 | 27 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.25 [0.74, 2.12] |
| 1.4 Live birth in women with more than two previous miscarriages | 1 | 119 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.00 [0.75, 1.33] |
| 2 Preterm delivery < 37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Preterm delivery 24‐28 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Preterm delivery 28‐32 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Preterm delivery 32‐37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Obstetric complications; pre‐eclampsia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Obstetric complications; IUGR Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 8 Congenital Malformations Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 9 Side effects; any bleeding Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 10 Side effects; pain/itching/swelling at injection site Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Live birth in all women | 2 | 327 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.11 [0.94, 1.30] |
| 1.2 Live birth in women with no previous live birth | 2 | 200 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.88, 1.31] |
| 1.3 Live birth in women with inherited thrombophilia | 2 | 63 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.05 [0.73, 1.50] |
| 1.4 Live birth in women with more than two previous miscarriages | 1 | 112 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.06 [0.78, 1.43] |
| 2 Preterm delivery < 37 weeks Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 3 Preterm delivery 24‐28 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Preterm delivery; 28‐32 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Preterm delivery; 32‐37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Obstetric complications; pre‐eclampsia Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Obstetric complications; IUGR Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 8 Congenital malformations Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 9 Side effects; any bleeding Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 10 Side effects; pain/itching/swelling at injection site Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 11 HIT/thrombocytopenia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Live birth in all women | 1 | 126 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.91 [0.72, 1.15] |
| 1.2 Live birth in women with no previous live birth | 1 | 72 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.77 [0.59, 1.02] |
| 1.3 Live birth in women with inherited thrombophilia | 1 | 32 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.48, 1.28] |
| 2 Preterm delivery < 37 weeks Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Obstetric complications; pre‐eclampsia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 4 Obstetric complications; IUGR Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 5 Congenital malformations Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 6 Side effects; any bleeding Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 5 | 793 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.99, 1.15] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Live birth Show forest plot | 3 | 324 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.85, 1.12] |
