Summary. The risks of long-term antenatal subcutaneous heparin therapy were assessed in a small controlled trial of prophylaxis of thromboembolism. Forty patients with a documented history of previous thromboembolism were randomly allocated either to receive heparin (10 000 i.u. subcutaneously twice daily) throughout pregnancy and labour or to receive no treatment (control group). All patients were treated with heparin (8000 i.u. twice daily) for 6 weeks after delivery from the first postnatal day. There appeared to be no increased risk of antenatal or postnatal bleeding associated with subcutaneous heparin, but one patient in the control group developed a deep vein thrombosis and one in the treatment group developed severe debilitating osteopenia. The withholding of epidural analgesia may have contributed to both maternal and fetal morbidity in the treatment group. There was one abortion in each group but no other fetal or neonatal losses although more babies from the treated group entered the special care baby unit. Although the numbers are too small for statistical analysis, the findings indicate that the use of long-term low-dose subcutaneous heparin is not without complications and there is need for a larger, multicentre trial to allow precise quantification of fetal and maternal risks against the risk of recurrent thromboembolism.