Thrombocytopenia is common in critically-ill patients, occurring in up to half of all patients in the intensive care unit (ICU). In this population, the presence of thrombo-cytopenia is associated with increased mortality, and depending on severity and etiology, is associated with increased hemorrhagic risk as well. ICU patients often have several potential causes of thrombocytopenia, making evaluation challenging. Heparin exposure (in the form of line flushes, prophylaxis, or therapeutic anticoagulation) is virtually ubiquitous in the ICU, making HIT a frequent diagnostic consideration.
There have been several prospective and retrospective studies that have evaluated the frequency of HIT in critically ill patients (Table 4). The two largest prospective studies (Verma et al., 2003; Crowther et al., 2005) found that although HIT was clinically suspected in 12 to 15% of ICU patients, compatible serology supporting the diagnosis (including the washed platelet activation assay, a relatively specific test for HIT antibodies) was found infrequently, with the overall frequency of HIT less than 1%. Indeed, if the frequency of HIT is only about 0.3-0.5% in ICU patients, yet the overall risk of thrombocytopenia is about 30-50%, then a useful clinical "rule" is that the true risk of HIT is only about one in 100 among all ICU patients who develop thrombocytopenia (Warkentin, 2006a; Napolitano et al., 2006; Selleng et al., 2007).
Other retrospective analyses utilizing anti-PF4/polyanion EIAs (which are much less specific for clinical HIT) raise the possibility that certain patient subgroups may be at higher risk of HIT. For example, Hoh et al. (2005) reported a much higher seropositivity rate in patients with subarachnoid hemorrhage admitted to a neurosurgical ICU who were suspected of having HIT on clinical grounds. Schmugge et al. (2002) claimed that 2.3% of critically ill pediatric patients exposed to heparin for at least 5 days developed HIT with thrombosis. Prospective studies utilizing platelet activation assays are required to ascertain the true frequencies of clinical HIT among these and other patient subgroups.
When HIT does occur in critically ill patients, it appears to be associated with a high risk of venous and arterial thrombotic events, similar to that observed in non-critically ill patients with HIT. Retrospective studies report a thrombosis frequency of 20-50%, or even greater (Wester et al., 2004; Hoh et al., 2005; Gettings et al., 2006). The HIT-associated mortality in these studies also appears to be high.
Was this article helpful?