A survey was carried out of the attitudes adopted in French laboratories with regard to the diagnosis of heparin-induced thrombocytopenia (HIT). The platelet aggregation assay is used in 100% of laboratories, aggregation being measured by light transmission in an aggregometer (3/4). Blood is drawn either in emergency (1/4) or after heparin discontinuation (1/4). The nature of the samples for testing is a fresh citrated plasma (100%) although frozen plasma is occasionally employed, while 40% of laboratories use less than 3 control platelet donors. Platelet response is verified in the presence of a non immune agonist (18%) or an immune challenge (known positive plasma or platelet activating monoclonal antibody) (13%). Heparin is of the same type as received by the patient and is tested at two or more concentrations of approximately 0.5 and 1.0 IU/ml, but rarely at high concentration (100 IU/ml). The platelet count is adjusted to 250 - 350 x 10(9)/l (18%), the ratio of patient plasma to PRP is 1:1 (59%), the time of observation is about 20 min (50%) and control platelets are tested with heparin to rule out any false positive results (9/10). Standardization is nevertheless required if the platelet aggregation assay is to be considered as a reference test.

Fibrin D-Dimer (D-Di), prothrombin activation fragment (F 1 + 2) and thrombin-antithrombin III complexes (TAT) were measured using ELISA procedures in the plasma of patients with an acute deep venous thrombosis (DVT), at presentation and on days 2, 6 and 10 after initiation of heparin treatment. Patients were randomly allocated into two treatment groups: 44 patients received adapted doses of continuous intravenous unfractionated heparin (UH) whereas 47 received 1 mg/kg every twelve hours of a low molecular weight heparin (enoxaparin) subcutaneously. A phlebography and a perfusion lung scan were performed before inclusion and on day 10. Failure of therapy (n = 9) was defined by venogram worsening or confirmed pulmonary embolism. Improvement (n = 44) or stationary state (n = 38) were defined by venogram evolution in the absence of new leg scan defects. At presentation, D-Di, F 1 + 2 and TAT were above cut-off values in 97, 66 and 89% of patients respectively. D-Di levels correlated with the extent of venous thrombosis whereas TAT and F 1 + 2 did not. Mean levels of D-Di decreased sharply during the first days of treatment but were still abnormal on day 10. A secondary increase of D-Di on days 6 or 10 by more than 3 mu g/ml occurred in 4 of the 9 patients who developed a thromboembolic recurrence but in none of the 72 patients who had a more favorable outcome. F 1 + 2 and TAT time-courses were not related to clinical evolution. In the Enoxaparin group, there was no relationship between antifactor Xa activities and any biological markers. TAT and F 1 + 2 levels fell on day 2 and remained stable until day 10. In contrast, in the UH group, TAT and F 1 + 2 did not significantly decrease on day 2, probably due to a delay in dose adaptation, but they declined slowly until day 10.