We report the case of 64-year-old female patient with pulmonary embolism and bilateral femoropopliteal deep vein thrombosis caused by heparin-induced thrombocytopenia type II (HIT II) resistant to danaparoid sodium and subsequently administered lepirudin in whom a single late plasmapheresis performed on day 6 of the initiation of treatment of HIT reversed the course of the disease, preventing its highly potential fatal outcome. Primarily administered lepirudin was not only ineffective but even led to further aggravation of the patient's clinical state and platelet count drop in the first stage of the HIT treatment. The improvement of the patient's clinical state was not achieved before therapeutic plasma exchange (TPE) had removed the greatest part of pathogenetic circulating substrate. Only after TPE, lepirudin, introduced again, led to the platelet count recovery. In the subsequent course of the treatment, lepirudin was combined with an overlapping oral anticoagulant. Previously positive heparin aggregation test and fast particle gel heparin-platelet factor 4 immunoassay were normalized as well as the patient's clinical status. Early plasmapheresis, administered within 4 days of the onset of thrombocytopenia in HIT, as a beneficial therapeutic measure in certain individual cases, is indisputable. However, our results do not concur with previously reported findings of the so far most comprehensive study on plasmapheresis performed in the management of HIT with thrombosis, discrediting late plasmapheresis administered 4 days after the onset of the disease not only as ineffective, but even as an aggravating factor. Our results suggest the possible beneficial impact of late plasmapheresis as a method that may reverse a prothrombotic process and lead to a fast improvement in the patient's platelet count, especially in cases initially resistant to thrombin inhibitors.