Risk factors for thrombosis in patients with immune mediated heparin-induced thrombocytopenia

Background. As reported by major clinical series in the literature, about 2% of patients receiving unfractionated heparin (UFH) develop immune-mediated (type II) heparin-induced thrombocytopenia (HIT) that may be complicated in 30-75% of cases by a paradoxical thrombotic syndrome (HITTS), either arterial or venous. HITTS carries relevant rates of mortality and morbidity, amongst which cerebral and/or myocardial infarction and limb amputations. It is unclear as yet why some patients suffer from isolated thrombocytopenia (HIT), whilst others have HITTS. The aim of the present study was to look for clinical and laboratory features related to the occurrence of HITTS. Patients and methods. We retrospectively analysed the clinical records of 56 patients with proven HIT, as diagnosed on clinical grounds and by in vitro demonstration of immunoglobulin (IgG)/IgM against the PF4/heparin complex. Thirty-four patients (61%) had HITTS (19 venous thrombosis, seven arterial thrombosis, five arterial and venous thrombosis, two skin necrosis, one diffuse intravascular coagulation), whereas 22 had uncomplicated HIT. Amongst HITTS patients, two had limb amputation, five had recurrent thrombosis and seven died. Amongst HIT patients three died from causes unrelated to HIT. Results. No significant difference in sex, age, previous exposure to heparin, UFH route of administration or dose, duration of therapy, time of onset of thrombocytopenia and platelet count recovery, nor antiheparin/PF4 antibodies subtype (IgG or IgM) was detected when comparing HIT and HITTS. In contrast, in the HITTS group a higher prevalence of orthopaedic surgery (15 of 34 vs. 2/22; P =0.01), a significantly lower platelet count nadir (43 +/- 32 vs. 75 +/- 63 x 10(9) /L; P =0.01) and a significantly higher titre of antiheparin/PF4 antibodies, expressed as optical density of enzyme-linked immunosorbent assay (ELISA); (1989 +/- 1024 vs. 1277 +/- 858; P =0.009), were observed in comparison with the HIT group. Amongst HITTS patients, the prevalence of venous thrombosis was significantly higher in orthopaedic patients and in those being treated for venous thromboembolism (18/24 vs. 1/9 patients, chi(2) 8.4, P =0.004), whilst arterial thrombosis (ART) occurred more often in heparin treatment for arterial disease (3/4 vs. 4/29 patients, chi(2) 4.6, P =0.03). Conclusions. Orthopaedic surgery, the severity of thrombocytopenia and high antiheparin/PF4 antibodies titre are adverse prognostic or concurrent factors in the development of HITTS.