Predictors of clinical outcome in patients with heparin-induced thrombocytopenia treated with direct thrombin inhibition.

We aimed to identify predictors of poor outcome in patients with heparin-induced thrombocytopenia, a serious immune-mediated reaction to heparin. All patients were treated with direct thrombin inhibition therapy, as part of two prospective studies. We performed a risk factor analysis of adverse outcomes (defined as death, amputation, new thrombosis, or their composite within a 37-day study period) in 809 patients from two reported prospective studies of the direct thrombin inhibitor argatroban in clinically diagnosed heparin-induced thrombocytopenia. We initially identified from among 14 baseline variables the significant predictors of poor outcome in the first study (304 patients), and then tested our resultant hypothesis in the second, independent study (505 patients), using multivariate analysis. Seven significant predictors were identified in the first study; three were confirmed in the second study. The strongest relationship occurred between the baseline platelet count and the composite of death, amputation, or new thrombosis (P = 0.0001), with the most severely thrombocytopenic patients being at greatest risk. The other significant associations were between renal impairment and death (odds ratio = 2.13, 95% confidence interval = 1.23-3.66, P = 0.007), and between cardiovascular surgery (particularly peripheral vascular surgery) and amputation (odds ratio = 3.39, 95% confidence interval = 1.65-6.95, P = 0.0009). In conclusion, in patients with clinically diagnosed heparin-induced thrombocytopenia, the severity of the baseline thrombocytopenia is the best predictor of death, amputation or thrombotic progression. The identification of higher risk subgroups for poor outcomes, such as patients with more severe thrombocytopenia or a history of renal impairment or peripheral vascular surgery, could allow more targeted therapy.