LAUSR

OBJECTIVES To investigate the independent association of platelet transfusion with hospital mortality and key relevant clinical outcomes in cardiac surgery. DESIGN A single-center, propensity score-matched, retrospective, cohort study. SETTING At an American tertiary teaching hospital data from the Medical Information Mart for Intensive Care III and IV databases from 2001 to 2019. PARTICIPANTS Consecutive adults undergoing coronary artery bypass graft and/or cardiac valvular surgery. INTERVENTIONS Platelet transfusion during perioperative intensive care unit (ICU) admission. MEASUREMENTS AND MAIN RESULTS Overall, 12,043 adults met the study inclusion criteria. Of these, 1,621 (13.5%) received apheresis-leukoreduced platelets, with a median of 1.19 units per recipient (IQR: 0.93-1.19) at a median of 1.78 hours (IQR: 0.75-4.25) after ICU admission. The platelet count was measured in 1,176 patients (72.5%) before transfusion, with a median count of 120 × 109/L (IQR: 89.0-157.0), and only 53 (3.3%) had platelet counts below 50 × 109/L. After propensity matching of 1,046 platelet recipients with 1,046 controls, perioperative platelet transfusion carried no association with in-hospital mortality (odds ratio [OR]: 1.28; 99% CI: 0.49-3.35; p = 0.4980). However, it was associated with a pattern of decreased odds of suspected infection (eg, respiratory infection, urinary tract infection, septicaemia, or other; OR: 0.70; 99% CI: 0.50-0.97; p = 0.0050), days in the hospital (adjusted mean difference [AMD]: 0.86; 99% CI: -0.27 to 1.98; p = 0.048), or days in intensive care (AMD 0.83; 99% CI: -0.15 to 1.82; p = 0.0290). CONCLUSIONS Platelet transfusion was not associated with hospital mortality, but it was associated with decreased odds of suspected infection and with shorter ICU and hospital stays.