Women who are pregnant and have mechanical prosthetic heart valves (MPHV) pose a major therapeutic dilemma for clinicians responsible for providing care. Two issues seem clear. Pregnancy in women with MPHV is very uncommon in the UK-the UKOSS study reports an occurrence in 0.0037% of pregnancies-one case in every 27 000 maternities, so very few UK practitioners will have direct experience in caring for these women. Pregnancy is a hypercoagulable state and women with MPHV require therapeutic dose anticoagulation throughout pregnancy to reduce the risk of valve thrombosis and its complications. A highly contentious issue is what constitutes the 'optimal' anticoagulant regi-men. Some clinicians (Regitz-Zagrosek et al. Eur Heart J 2011;32:3147-97) strongly favour continuing vitamin K antagonists, such as warfarin, throughout pregnancy, arguing that they are the most effective, albeit not 100% effective, at preventing thromboem-bolic complications (TEC). However, these drugs freely cross the placenta and are teratogenic. Warfarin embry-opathy is reported in up to 12% of infants exposed in the first trimester. Warfarin fetopathy (fetal loss, stillbirth and neurological problems) is reported in up to 25% women exposed to war-farin in later pregnancy. Much has been made of a 'safe' warfarin dose but there is no dose that prevents adverse fetal outcomes (McLintock Best Pract Res Clin Obstet Gynecol 2014;28:519-36). This has prompted the demand for alternative approaches to anticoag-ulation. Low-molecular-weight heparin (LMWH) does not cross the placenta, is not teratogenic and its use is associated with live birth rates of 95% in women with MPHV. Although it is not as effective at preventing TEC, the best outcomes are achieved if women are able to comply with twice daily injections of LMWH and have regular anti-Xa levels to guide dose adjustments as pregnancy progresses. Critical to the clinical efficacy of LMWH is ensuring adequate levels of anticoagu-lation. Peak anti-Xa levels will predict the maximum anticoagulant effect of LMWH and risk of bleeding. Trough levels ensure that anticoagulation is within a therapeutic threshold. Whether targeting peak and trough levels is associated with lower risks of TEC has not been tested in a large prospective study. The UKOSS study highlights an inconsistent and inadequate approach to anticoagulant management. Practitioners did not seem to appreciate the complexities inherent in providing safe anticoagulation or the high risk of these pregnancies. One in five women were not referred for specialist care at any stage during their pregnancy. Of the women who chose 'therapeutic dose' LMWH throughout pregnancy, data on anti-Xa monitoring was available in just over half, with alarming variations in the frequency and timing of testing, and target anti-Xa levels. Inappropriate LMWH dosing and poor compliance were associated with poor maternal outcome. Given the inadequacies in clinical care, it is not surprising that clinical outcomes in the cohort overall were poor and it is uncertain if...

There is a pressing need for further research to identify the specific concerns of late bookers, to identify areas where new interventions might encourage the uptake of services and to gauge the likely impact of increased dissemination of information about the availability of antenatal care services.