Blood transfusion in coronary artery surgery: focus on modifiable risk factors

Abstract

We read with great interest the recently published study by De Santo et al. [1]. The study was conceived to identify preoperative and intraoperative patient characteristics predicting a higher risk of red blood cell (RBC) transfusion in isolated coronary artery bypass grafting (CABG), in order to reveal factors or practices that might be modified [1]. Eight predictors emerged through regression logistic analysis: age, body surface area, preoperative glomerular filtration rate, preoperative haemoglobin, surgical priority, length of cardiopulmonary bypass (CPB), intraoperative haemodilution and early postoperative blood loss [1]. The transfused group had higher values of chest tube output (CTO), P < 0.0001 [1]. CTO presents typical modifiable factors, but the question ‘How to predict or prevent excessive CTO?’ remains challenging. In our opinion, when seeking for modifiable risk factors for blood transfusion, both preand intraoperative objective quantification of platelet activity as well as assessment of viscoelastic blood clot properties using rotational thromboelastometry should inextricably be taken into consideration [2, 3]. The efficacy of platelet inhibition with aspirin (ASA) and clopidogrel (CLO) varies widely among patients, from intensive platelet inhibition to poor platelet response [4], and those facts could, to a certain degree, explain no impact of CLO and ASA administration on transfusion outcome [1]. Notably, it remains unclear how many patients were exposed to dual antiplatelet therapy (DAT) with ASA + CLO preoperatively. Was proportion of patients preoperatively exposed to DAT similar between the transfused and non-transfused groups? In addition, transfused patients more frequently underwent emergent or urgent surgery. Were patients in emergent and/or urgent subgroup more frequently exposed to DAT? The possible role of DAT in assessment of transfusion outcome should not be underestimated since further incremental platelet inhibition may be observed in the group of patients receiving DAT [5]. In our experience, prediction of excessive CTO is possible both preand intraoperatively [2, 3]. Recently, we found ASA(P = 0.014) and CLO(P = 0.003) sensitive platelet function tests to be predictive of excessive CTO in patients following CABG [2]. One hundred and sixty-one (76.3%) patients received RBC with no significant differences in RBC administration among the groups with regard to preoperative antiplatelet drug administration regime (P = 0.636) [2], which is in line with results in the present study [1]. However; comparison of the ASA-sensitive platelet function test values between patients with respect to packed red blood cells administration revealed significantly lower test values in the group of patients exposed to RBC (P = 0.002) [2]. The role of ASA and CLO administration management should be separately assessed by drug-specific platelet function tests, thus facilitating an individual therapeutic approach for each antiplatelet agent preoperatively. In addition, intraoperative assessment of platelet function and viscoelastic blood clot properties during CPB can reveal a further degree of haemostatic disorder and its relation to bleeding extent as well as transfusion requirements [3]. Preand intraoperative assessment of platelet function and viscoelastic blood clot properties can distinguish the influence of pre-existing, antiplatelet drugsrelated and CPB-acquired haemostatic disorders, allowing detection of risk factors and enabling preoperative (procedure timing, risk stratification, antiplatelet therapy discontinuation management) and intraoperative (targeted administration of desmopressin, tranexaminic acid and procoagulant blood components) practice modifications, which may further lead to improvement in transfusion as well as bleeding, and thus clinical outcome.