The debate goes on. Antithrombotic prophylaxis for women with thrombophilia and pregnancy complications is controversially discussed. In the October 2003 issue of the Journal of Thrombosis and Haemostasis, Middeldorp argued against the implementation of a Ôpotential harmful therapyÕ in women with hereditary thrombophilia and a history of recurrent pregnancy loss [1]. In the same issue, Brenner argued in favor of a heparin prophylaxis, based on the results of the limited number of studies available [2]. A forum on the subject in the July 2004 issue reflected the controversial positions [3][4][5][6][7]. Now, 1 year later, another forum on lower-molecular-weight heparin (LMWH) prophylaxis has been opened following a recent publication of Brenner et al. Since the first debate, an important prospective study has been published supporting the use of LMWH. Gris et al. [11] documented the superiority of LMWH when compared with aspirin in the prevention of fetal loss in women with a thrombophilic defect after a first pregnancy loss. Although this report did not deal with recurrent pregnancy loss, the study further supports the benefit of LMWH in women at risk. The recent trial by Brenner et al., comparing enoxaparin 40 and 80 mg day )1 , showed no significant difference in clinical outcome, indicating that the maximum effect of heparin treatment is already obtained with 40 mg day )1 [8]. A limitation of Brenner's study is the lack of a control group without heparin prophylaxis. Therefore, no validated conclusion on a treatment effect can be drawn.Although data in favor of LMWH are increasing, there are only a limited number of studies supporting heparin prophylaxis in women with recurrent pregnancy loss. Most of these studies are of low quality because of their historic controls. Although LMWH is widely used in pregnancy for prophylaxis of venous thromboembolism and although relevant adverse reactions are rare [12], its application in women with fetal loss requires thorough information on the risk-benefit ratio.There is much discussion on the impact of thrombophilic factors as risk determinants of pregnancy complications including recurrent pregnancy loss or pre-eclampsia [10]. In contrast to venous thromboembolism, the data are more conflicting, particularly in women with early fetal loss. We agree with Lindqvist et al. who, in their recent comment, question the validity of meta-analyses on this issue [10]. These meta-analyses are based on small retrospective studies, which are subject to a publication bias. More importantly, study results are likely to be biased by the increasing referral of women with predetermined thrombophilia to centers which enroll these patients in case-control studies. In consequence, the resulting odds ratio of thrombophilia is overestimated and reflects primarily a referral bias. As the placenta becomes the essential source of blood supply to the embryo between 8 and 10 weeks of gestation [9,13], early and late fetal loss must be separately evaluated and the enrollment of patients has to be ca...