337 MANAGEMENT OF SEVERE CARDIOGENIC SHOCK WITH THE IMPELLA MICROAXIAL LEFT VENTRICULAR ASSIST DEVICE: EARLY EXPERIENCE IN A SINGLE CENTER A Mazine, J Forcillo, P Demers, D Bouchard, Y Lamarche, M Carrier Montréal, Québec BACKGROUND: Despite recent advances in therapeutic interventions and improvements in medical care, cardio- genic shock is still associated with high mortality. The Impella left ventricular assist device is a microaxial pump designed to unload the left ventricle and provide temporary support in the setting of severe left ventricular dysfunction. The purpose of this study is to report our initial experience with the Impella device in patients with severe cardiogenic shock. METHODS: Between October 2010 and June 2012, 16 patients with acute, refractory cardiogenic shock underwent implantation of an Impella left ventricular assist device. The LP 2.5, LD 5.0 and LP 5.0 models were used in 5 (31%), 1 (6%) and 10 (63%) patients, respectively. The etiology of cardiogenic shock was ST-elevation myocardial infarction in 6 (38%) patients, post-cardiotomy low output syndrome in 2 (13%) patients and non-ischemic cardio- myopathy in 8 (50%) patients. Pre-implantation and post- implantation hemodynamic values were obtained for each patient. RESULTS: Mean age was 51.9 Æ 14.6 years and 12 (75%) patients were male. Median NYHA functional class was IV/IV. Prior to implantation, all patients were on iono- tropic support, 11 (69%) were mechanically ventilated, 5 (31%) had received an intra-aortic balloon pump and 4 (25%) had suffered from ventricular fibrillation or undergone cardiopulmonary resuscitation. After Impella implantation, there was significant improvement of the left ventricular ejection fraction (19 Æ 8% vs 31 Æ 20%, p ¼ 0.015) and of the cardiac index (1.6 Æ 0.4 L/min/m 2 vs 3.0 Æ 0.7 L/min/m 2+ , p ¼ 0.008). Mean duration of circulatory support was 93 Æ 70 hours. In-hospital mortality occurred in 5 (31%) patients, including 3 who had ST-elevation myocardial infarction and 2 who had non-ischemic cardiomyopathy. The remaining 11 (69%) patients were successfully weaned from the Impella device. Of these, six were discharged, one underwent percutaneous coronary intervention, one underwent implantation of a ventricular assist device and three received a heart transplantation. CONCLUSION: This study suggests that the Impella device may be useful in reducing mortality in patients with severe refractory cardiogenic shock. Use of this microaxial pump seems to be particularly beneficial in patients with non- ischemic shock. However, further studies are needed to define its indications more clearly. 338 DIFFERENTIATION OF HEPARIN AND NON-HEPARIN INDUCED THROMBOCYTOPENIA THROUGH EARLY PLATELET MEASURES A Villafranca, R Manji, IA Thomson, B Arenson, J Lucht, E Jacobsohn Winnipeg, Manitoba BACKGROUND: Heparin induced thrombocytopenia (HIT) occurs in 1-3% of patients following cardiac surgicy and is associated with significant morbidity and mortality. Differ- entiation between HIT and other causes of thrombocyto- penia in the early postoperative period can be difficult, since: the gold standard test for HITT (serotonin release assay) is not widely available, while the more widely available ELISA test is oversensitive. Currently postoperative patients with HIT may be discharged from hospital before showing symptoms. We therefore sought to determine whether differences in platelet curves during postoperative days 1-4 could promote early differentiation of HITT from other TCP. METHODS: We performed a retrospective chart review of 94 patients who underwent cardiac surgery and had suspected HIT, of who 15 were confirmed by SRA. Each patient's platelet counts were interpolated. Two time variables were used to model the observed reverse para- bolic platelet-time curve: time since surgery and time since postoperative day 2. A series of mixed models were developed. RESULTS: Predictors of platelet count changes in a given person (fixed effects) included: time measures, HIT test S226 Canadian Journal of Cardiology Volume 29 2013