RESUMEN
OBJECTIVE: The prognostic impact of thrombocytopenia in patients supported by extracorporeal membrane oxygenation after cardiac surgery is uncertain. We investigated whether thrombocytopenia is independently predictive of poor outcome and describe the incidence and time course of thrombocytopenia in extracorporeal membrane oxygenation patients. DESIGN: Retrospective analysis of prospectively collected data. SETTING: Cardiosurgical ICU at a tertiary referral center. PATIENTS: Three hundred adult patients supported with venoarterial extracorporeal membrane oxygenation for more than 24 hours because of refractory cardiogenic shock after heart surgery between January 2001 and December 2014. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Two-way analysis of variance was used to compare the time course of platelet count changes between survivors and nonsurvivors. Using multiple Cox regression with time-dependent covariates, we investigated the impact of platelet count on 90-day mortality. In nonsurvivors, the daily incidence of moderate (< 100 - 50 × 10/L), severe (49 - 20 × 10/L), and very severe (< 20 × 10/L) thrombocytopenia was 50%, 54%, and 7%, respectively. Platelet count had a biphasic temporal pattern with an initial decrease until day 4-5 after the initiation of extracorporeal membrane oxygenation. Although a significant recovery of the platelet count was observed in survivors, a recovery did not occur in nonsurvivors (p = 0.0001). After adjusting for suspected confounders, moderate, severe, and very severe thrombocytopenia were independently associated with 90-day mortality. The highest risk was associated with severe (hazard ratio, 5.9 [2.7-12.6]; p < 0.0001) and very severe thrombocytopenia (hazard ratio, 25.9 [10.7-62.9], p < 0.0001). CONCLUSION: Thrombocytopenia is an independent risk factor for poor outcome in extracorporeal membrane oxygenation patients after cardiac surgery, with persistent severe thrombocytopenia likely reflecting a high degree of physiologic imbalance.