Since the recognition of the antiphospholipid syndrome, a great number of cardiac manifestations have been reported in association with these antibodies: valvular disease, coronary artery disease, cardiomyopathy and intracardiac thrombosis. However this association raises numerous questions related to the pathogenic role of antiphospholipids, their prognostic significance and their frequency in a non-selected population with a definite cardiac manifestation. In view of the literature and our personal experience, it seems necessary to distinguish two kinds of situations. During systemic lupus and primary antiphospholipid syndrome (which must be systematically looked for in patients with history of thrombo-embolic disease), antiphospholipids antibodies certainly play a role in the occurrence of cardiac manifestations, but the precise place of thrombosis has to be best defined along with immunologic/inflammatory mechanisms. On the other hand, in a non-selected population, antiphospholipids antibodies may just be the consequence of the cardiac lesion and do not seem to have prognostic implications. This distinction, actually hypothetical, should be supported on the basis of distinct specificities of antiphospholipids antibodies and especially their dependence on beta 2-glycoprotein I, which would help to distinguish the harmful antibodies from those which probably just appear as an epiphenomenon.