Platelet transfusions in platelet consumptive disorders are associated with arterial thrombosis and in-hospital mortality

Abstract

While platelets are primary mediators of hemostasis, there is emerging evidence to show that they may also mediate pathologic thrombogenesis. Little data are available on risks and benefits associated with platelet transfusions in thrombotic thrombocytopenic purpura (TTP), heparin-induced thrombocytopenia (HIT) and immune thrombocytopenic purpura (ITP). This study utilized the Nationwide Inpatient Sample to evaluate the current in-hospital platelet transfusion practices and their association with arterial/venous thrombosis, acute myocardial infarction (AMI), stroke, and in-hospital mortality over 5 years (2007-2011). Age and gender-adjusted odds ratios (adjOR) associated with platelet transfusions were calculated. There were 10 624 hospitalizations with TTP; 6332 with HIT and 79 980 with ITP. Platelet transfusions were reported in 10.1% TTP, 7.1% HIT, and 25.8% ITP admissions. Platelet transfusions in TTP were associated with higher odds of arterial thrombosis (adjOR = 5.8, 95%CI = 1.3-26.6), AMI (adjOR = 2.0, 95%CI = 1.2-3.3) and mortality (adjOR = 2.0,95%CI = 1.3-3.0), but not venous thrombosis. Platelet transfusions in HIT were associated with higher odds of arterial thrombosis (adjOR = 3.4, 95%CI = 1.2-9.5) and mortality (adjOR = 5.2, 95%CI = 2.6-10.5) but not venous thrombosis. Except for AMI, all relationships remained significant after adjusting for clinical severity and acuity. No associations were significant for ITP. Platelet transfusions are associated with higher odds of arterial thrombosis and mortality among TTP and HIT patients.

Figure 1

Multivariable OR and 95% CI of platelet transfusions and various comorbidities/in-hospital mortality in hospitalized patients with TTP, HIT, and ITP. Data are representative of a weighted nationally representative sample of US community hospitals across 5 years (NIS, 2007 to 2011).