Summary. Heparin-induced thrombocytopenia and/or thrombosis (HITT) are serious complications of heparin treatment. The incidence, as previously reported, varies widely and, in consequence, is not precisely known. Moreover, most reports only concern clinically defined heparininduced thrombocytopenia. Therefore we carried out a prospective study of the incidence of serologically confirmed HITT.All patients admitted to the Departments of Cardiology and Neurology of our institution with an indication for treatment with therapeutic-dose intravenous unfractionated heparin were enrolled in the study. The patients were examined daily for the occurrence of thromboembolic complications. Regular platelet counts and tests for the presence of heparin-dependent antibodies were carried out using two different tests: a quantitative platelet factor 4/ heparin (PF4/hep) Elisa, and a functional test, the heparininduced platelet activation assay (HIPAA). HITT was defined as a rapidly occurring (within 5 d) decrease of the platelet count from normal values of > 120 × 10 9 = l to < 60 × 10 9 = l or to < 100 × 10 9 = l if there was a rapid fall of >50% of starting value or >30% with concomitant acute thrombosis.The observed incidence of HITT was 1/358 patients (0·3%, 95% confidence limits 0·01-1·5%). However, Elisa PF4/hep specific IgG antibodes were demonstrated in nine (2·5%) and IgM antibodies in seven (2·0%) of 358 patients. 30/358 patients (8·4%) had platelet activating antibodies in the HIPAA.We conclude that the incidence of serologically confirmed HITT in this study is very low (0·3%) in patients with cardiac and neurologic diseases treated with intravenous unfractionated heparin. The frequency of heparin-dependent antibodies without concomitant occurrence of thrombocytopenia is much higher.