We read with great interest the recently published multicentre cohort study by Kremke et al. [1]. Patients exposed to antiplatelet therapy (APT) at the time of coronary artery surgery (CAS) had greater postoperative bleeding volumes and greater transfusion requirements [1]. Exposure to clopidogrel, but not aspirin, was associated with greater reoperation rates and was an independent risk factor for severe postoperative bleeding, although mean bleeding volumes were not significantly different in the aspirin and clopidogrel subgroups [1]. Despite current guidelines, it is apparent that many patients are proceeding with CAS without the 5-day delay off clopidogrel. The efficacy of platelet inhibition with clopidogrel varies widely among patients, from intensive platelet inhibition to poor platelet response [2], and these variations could, to some degree, explain the non-significant differences in mean bleeding volumes in the aspirin and clopidogrel subgroups [1]. The effect of clopidogrel on bleeding mainly depends on two factors: (i) observed platelet inhibition, which depends on inherent platelet activity prior to clopidogrel administration and platelet inhibitory response to clopidogrel and (ii) newborn platelets ability to restore normal aggregation after clopidogrel discontinuation. Therefore, the use of suitable point-of-care platelet function analysers seems to be reasonable in this field. Recently, we found aspirin(P = 0.014) and clopidogrel(P = 0.003) sensitive platelet function tests to be predictive of excessive postoperative bleeding in patients following CAS [2]. Of note, 76.3% patients were transfused with no significant differences among the groups with regard to the preoperative APT administration regime (P = 0.636) [2]. However, transfused patients had significantly lower values of aspirin-sensitive platelet function tests (P = 0.002) [2]. In our opinion, the study cohort is somehow heterogeneous. Eligible procedures were either isolated CAS or CAS combined with aortic valve replacement (AVR) [1]. AVR certainly extends the duration of cardiopulmonary bypass, which was recently described as a predictor of transfusion requirements [3]. After matching, authors reported a slightly greater use of antifibrinolytic drugs in the APT group [1]. Both on-pump and off-pump procedures were included [1]. Authors reported that off-pump surgery and perioperative use of tranexamic acid were associated with a lower risk of severe postoperative bleeding [1]. The use of antifibrinolytic drugs should be equally balanced since there is evidence that antifibrinolytic therapy reduces bleeding and transfusion outcomes [4]. It would be interesting to see if exclusion of CAS + AVR and off-pump CAS patients would bring different results. During the study period, thromboelastometry-guided blood component therapy became available at two of three participating centres [1]. At the same time, multivariate analysis revealed that surgery at one of the participating centres was one of independent risk factors for severe postoperative bleeding [1]. Spiess et al. [5] reported thromboelastography-guided haemostatic management to have significantly reduced the incidence of overall transfusion and mediastinal re-exploration due to excessive bleeding. Thus, perhaps one could assume that one centre, which was found to be independent risk factor for severe postoperative bleeding, was in fact the one without thromboelastometry guided transfusion therapy available. The largest extent of chest tube output together with the greatest transfusion rates, was observed in the subgroup of patients exposed to dual APT [1]. Those results are in line with the results recently published by Miceli et al. [6]. Further incremental platelet inhibition may be observed in the group of patients receiving dual APT. Therefore, the role of aspirin and clopidogrel administration management should be separately assessed by drug-specific platelet function tests, thus facilitating an individual therapeutic approach for discontinuation management of each antiplatelet agent preoperatively.
Read full abstract