Should we include anti-prothrombin antibodies in the screening for the antiphospholipid syndrome?

Anti-prothrombin antibodies belong to the family of the antiphospholipid antibodies. Tnt ir prevalence ranges from 50 to 90% of antiphospholipid-positive patients, depending on the laboratory methodology employed for their detection. ELISA techniques are the most commonly used methods for their measurements, which allow a quick determination of their titer and isotype(s). Unfortunately, no well standardized assays are yet commercially available. The clinical relevance of anti-prothrombin antibodies as risk factors for thromboembolic events has been investigated by a number of retrospective and case-control studies. We reviewed 11 studies: data on 1,440 patients and 371 sex- and age-matched controls were available. Seven studies reported a significant association between anti-prothrombin antibodies and thrombosis by univariate analysis. Multivariate analysis was performed in five studies: only two of them confirmed the association. The sensitivity and specificity of anti-prothrombin antibodies for thrombosis were analysed according to the antibody isotype and to the arterial and/or venous site of thrombosis. The sensitivities are disappointingly low, particularly when the M isotype and arterial thrombosis are considered, whereas the specificities are somewhat better, even though they range widely. These data do nut allow to recommend the measurement of anti-prothrombin antibodies in the routine laboratory workout of antiphospholipid-positive patients in order to define their thrombotic risk. (C) 2000 Academic Press.