Venous thromboembolism (VTE) occurs when a blood clot forms in a deep vein, usually in a leg, forming a deep venous thrombosis (DVT), which may cause pain and swelling. This is very rarely fatal, but if part of the clot breaks off it may be carried to the lungs by the blood and block vessels there (this is called a pulmonary embolism (PE)), which can result in death. Normally occurring changes to the clotting system during pregnancy can increase the risk of a thromboembolic event (DVT or PE), and some groups of women have a higher risk of developing VTE (including older and obese women; women with previous VTE; women with thrombophilia (a condition predisposing individuals to developing clots); and women following a caesarean birth). Preventive treatments include drugs to prevent clots, support stockings, and physical activity soon after birth to keep the circulation moving. However, some drugs might cause problems such as increased blood loss after birth. Drugs used include unfractionated heparin, low molecular weight heparin and aspirin.
We included 19 randomised controlled trials in this review but only 16 trials with 2592 women could be included in the analysis. The trials were of a moderate quality, and assessed drugs including unfractionated heparin and low molecular weight heparin in pregnancy and after caesarean birth. We found no evidence to suggest that using heparin in pregnancy or after a caesarean birth reduces the risk of maternal death, DVT or PE, and no differences were shown for these outcomes when different types of heparin were compared. Women who received low molecular weight heparin seemed to be less likely to have bleeding episodes (bruises of more than 1 inch; injection site haematoma (a localised collection of blood outside blood vessels) of at least 2 cm in diameter, bleeding at birth and other bleeding), were less likely to have injection site burning, excess bruising and allergic rashes, and less likely to have a fetal loss, than women who received unfractionated heparin; however, the trials that showed these results were not of high quality.
We did not find enough evidence from the trials to be sure about the effects of these preventive treatments. This means there is not enough evidence to show which are the best ways to prevent VTE (including DVT and PE), during or following pregnancy, including after a caesarean birth.
There is insufficient evidence on which to base recommendations for thromboprophylaxis during pregnancy and the early postnatal period, with the small number of differences detected in this review being largely derived from trials that were not of high methodological quality. Large scale, high-quality randomised trials of currently used interventions are warranted.
Venous thromboembolism (VTE), although rare, is a major cause of maternal mortality and morbidity, and methods of prophylaxis are therefore often used for women considered to be at risk. This may include women who have given birth by caesarean section, those with a personal or family history of VTE and women with inherited or acquired thrombophilias (conditions that predispose people to thrombosis). Many methods of prophylaxis carry risks of adverse effects, and as the risk of VTE is often low, it is possible that the benefits of thromboprophylaxis may be outweighed by harms. Guidelines for clinical practice have been based on expert opinion rather than high-quality evidence from randomised trials.
To assess the effects of thromboprophylaxis in women who are pregnant or have recently given birth and are at increased risk of VTE on the incidence of VTE and adverse effects of treatment.
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register (27 November 2013).
Randomised trials comparing one method of thromboprophylaxis with placebo or no treatment, and randomised trials comparing two (or more) methods of thromboprophylaxis.
At least two review authors assessed trial eligibility and quality and extracted the data.
Nineteen trials, at an overall moderate risk of bias, met the inclusion criteria for the review. Only 16 trials, involving 2592 women, assessing a range of methods of thromboprophylaxis, contributed data to the review. Six trials compared methods of antenatal prophylaxis: heparin versus no treatment/placebo (two trials), and low molecular weight heparin (LMWH) versus unfractionated heparin (UFH) (four trials). Nine trials assessed prophylaxis after caesarean section: four compared heparin with placebo; three compared LMWH with UFH; one compared hydroxyethyl starch (HES) with UFH; and one compared five-day versus 10-day LMWH. One study examined prophylaxis with UFH in the postnatal period (including following vaginal births).
For antenatal prophylaxis, none of the included trials reported on maternal mortality, and no differences were detected for the other primary outcomes of symptomatic thromboembolic events, symptomatic pulmonary embolism (PE) and symptomatic deep venous thrombosis (DVT) when LMWH or UFH was compared with no treatment/placebo or when LMWH was compared with UFH. The risk ratios (RR) for symptomatic thromboembolic events were: antenatal LMWH/UFH versus no heparin, RR 0.33; 95% confidence interval (CI) 0.04 to 2.99 (two trials, 56 women); and antenatal LMWH versus UFH, RR 0.47; 95% CI 0.09 to 2.49 (four trials, 404 women). No differences were shown when antenatal LMWH or UFH was compared with no treatment/placebo for any secondary outcomes. Antenatal LMWH was associated with fewer adverse effects sufficient to stop treatment (RR 0.07; 95% CI 0.01 to 0.54; two trials, 226 women), and fewer fetal losses (RR 0.47; 95% CI 0.23 to 0.95; three trials, 343 women) when compared with UFH. In two trials, antenatal LMWH compared with UFH was associated with fewer bleeding episodes (defined in one trial of 121 women as bruises > 1 inch (RR 0.18, 95% CI 0.09 to 0.36); and in one trial of 105 women as injection site haematomas of ≥ 2 cm, bleeding during delivery or other bleeding (RR 0.28; 95% CI 0.15 to 0.53)), however in a further trial of 117 women no difference between groups was shown for bleeding at delivery. The results for these secondary outcomes should be interpreted with caution, being derived from small trials that were not of high methodological quality.
For post-caesarean/postnatal prophylaxis, only one trial comparing five-day versus 10-day LMWH after caesarean section reported on maternal mortality, observing no deaths. No differences were seen across any of the comparisons for the other primary outcomes (symptomatic thromboembolic events, symptomatic PE and symptomatic DVT). The RRs for symptomatic thromboembolic events were: post-caesarean LMWH/UFH versus no heparin, RR 1.30; 95% CI 0.39 to 4.27 (four trials, 840 women); post-caesarean LMWH versus UFH, RR 0.33; 95% CI 0.01 to 7.99 (three trials, 217 women); post-caesarean five-day versus 10-day LMWH, RR 0.36; 95% CI 0.01 to 8.78 (one trial, 646 women); postnatal UFH versus no heparin, RR 0.16; 95% CI 0.02 to 1.36 (one trial, 210 women). For prophylaxis after caesarean section, in one trial (of 580 women), women receiving UFH and physiotherapy were more likely to have bleeding complications ('complications hémorragiques') than women receiving physiotherapy alone (RR 5.03; 95% CI 2.49 to 10.18). In two additional trials, that compared LMWH with placebo, no difference between groups in bleeding episodes (major bleeding; major bruising; bleeding/bruising reported at discharge) were detected. No other differences in secondary outcomes were shown when LMWH was compared with UFH post-caesarean, nor when post-caesarean HES was compared with UFH, post-caesarean five-day LMWH was compared with 10-day LMWH, or when UFH was compared to no heparin postnatally.