775

Abstract

Introduction: Heparin Induced Thrombocytopenia (HIT) in critically ill patients remains associated with increased morbidity and mortality. Yet, the burden of this disease in the intensive care unit (ICU) remains poorly understood. We sought to clarify the incidence of HIT in ICU patients. Hypothesis: HIT occurs infrequently in non-cardiac surgery ICU patients. Methods: We performed a prospective observational study in consecutive ICU patients who underwent HIT antibody testing (August 2009 – March 2012). The decision to pursue HIT diagnostically was left to the primary team. The enzyme immunoassay served as the initial screening test for HIT and by convention an optical density > 0.4 defined a positive test. The final diagnosis of HIT was confirmed by a serotonin release assay (SRA) which was performed in a central lab in all subjects. The incidence of HIT represented our primary endpoint. We then compared rates of HIT in various patient populations. Results: The cohort consisted of 218 subjects (mean age: 62.5 +/- 17.1; male gender: 56.0%). An SRA was positive and the diagnosis of HIT confirmed in 3.2% of patients tested. Physicians were more likely to order HIT in the surgical (SICU) compared to the medical service (67.5% vs 32.5%). HIT was diagnosed in 1.1% of the medical ICU patients tested vs 4.7% of the surgical ICU patients (p=0.17). Among the surgical population diagnosed with HIT, 83.3% had undergone cardiovascular surgery prior to the development of HIT. Coronary artery bypass grafting (CABG) was the most common cardiovascular surgery that predisposed to HIT (49.6%). In all non-cardiac surgery ICU patients, HIT occurred in 0.00%. Conclusions: HIT remains rare in the ICU, particularly in the general medical-surgical ICU patient population. The highest burden of this disease in the ICU is associated with those who undergo cardiovascular surgery. More importantly, patients who undergo CABG have higher risk than other general surgical ICU patients. Given the low prevalence of HIT in the non-cardiac ICU patient population, testing for HIT should only be pursued if the pre-test probability for HIT is high.