OBJECTIVE: To evaluate the role of anti-beta(2)-glycoprotein-I (anti-beta(2)GPI-ab) and anti-phosphatidylserine (aPS-ab) antibodies as a risk factor in both recurrent miscarriage (RM) and unexplained fetal losses (UFL). DESIGN: Retrospective, cohort study. SETTING: Vall d'Hebron University Hospital, Barcelona, Spain. PATIENT(S): 122 pregnant women divided in two groups: study group of 54 women with RM and/or UFL and control group of 68 pregnant without RM history. INTERVENTION(S): Analysis of lupus anticoagulant, anticardiolipin antibodies, and anti-beta(2)GP1 and aPS antibodies. MAIN OUTCOME MEASURE(S): Comparison of aPL antibody between groups. RESULT(S): The prevalence of aPL positive results was 8 out of 54 (14.8%) in the study group and 3 out of 68 (4.41%) in the controls. In the RM subgroup, the prevalence was 3 out of 25 (12%) versus 3 out of 68 (4.4%), and 7 out of 34 (20.6%) versus 3 out of 68 (4.4%) in UFL subgroup. As a whole, the prevalence of anti-beta(2)GP1-ab in the RM/UFL group showed a difference compared with controls but not aPS-ab. In the RM women, anti-beta(2)GP1-ab was positive in 3 out of 25 (12%) versus 1 out of 68 (1.5%) in controls and in 4 out of 34 versus 0 out of 68 cases in women with UFL. In the RM subgroup, aPS-ab was positive in 1 out of 25 (4%) versus 2 out of 68 (2.9%) in control group and in 3 out of 34 versus 2 out of 68 cases in women with UFL. CONCLUSION(S): Our results suggest that anti-beta(2)GP1-ab but not aPS-ab is related to RM/UFL and should be considered as a pregnancy-loss risk factor.