- Email: [email protected]
Does Platelet Reactivity Predict Bleeding in Patients Needing Urgent Coronary Artery Bypass Grafting During Dual Antiplatelet Therapy?
Does Platelet Reactivity Predict Bleeding in Patients Needing Urgent Coronary Artery Bypass Grafting During Dual Antiplatelet Therapy? Elisabeth Mahla, MD, Florian Prueller, MD, Sylvia Farzi, MD, Gudrun Pregartner, MSc, Reinhard B. Raggam, MD, Elisabeth Beran, MD, Wolfgang Toller, MD, Andrea Berghold, PhD, Udaya S. Tantry, PhD, and Paul A. Gurbel, MD Department of Anesthesiology and Intensive Care Medicine, Medical University of Graz, Graz; Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz; Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz; Department of Cardiac Surgery, Medical University of Graz, Graz, Austria; and Inova Center for Thrombosis Research and Drug Development, Inova Heart and Vascular Institute, Falls Church, Virginia
Background. Up to 15% of patients require coronary artery bypass grafting (CABG) during dual antiplatelet therapy. Available evidence suggests an association between platelet reactivity and CABG-related bleeding. However, platelet reactivity cutoffs for bleeding remain elusive. We sought to explore the association between platelet reactivity and bleeding. Methods. Patients on aspirin and a P2Y12 receptor inhibitor within 48 hours before isolated CABG (n [ 149) were enrolled in this prospective study. Blood was drawn 2 to 4 hours preoperatively and platelet reactivity assessed by light transmittance aggregometry (LTA), vasodilatorstimulated phosphoprotein (VASP) assay, Multiplate analyzer and Innovance PFA2Y. The primary endpoint was calculated red blood cell loss computed as follows: (blood volume 3 preoperative hematocrit 3 0.91) – (blood volume 3 hematocrit 3 0.91 on postoperative day 5) D (mL of transfused red blood cells 3 0.59). Results. Preoperative platelet reactivity was low [median (interquartile range): LTA: 20 (9–28)%; VASPPRI: 39 (15–73)%; Multiplate adenosine phosphate test: 16 (12–22) U*min]. Innovance PFA2Y ‡300 seconds,
72%. Median (IQR) red blood cell loss in patients in first the LTA tertile was 1,449 (1,020 to 1,754) mL compared with 1,107 (858 to 1,512) mL and 1,075 (811 to 1,269) mL in those in the second and third tertiles, respectively (p < 0.004). Bleeding Academic Research Consortium (BARC)-4 bleeding differed between tertiles (62% versus 46% versus 36%; p ¼ 0.037). In a multivariable linear regression model, aspirin dose ‡300 mg, cardiopulmonary bypass time, EuroSCORE, and tertile distribution of platelet reactivity were significantly associated with red blood cell loss. Conclusions. A gradual decrease in red blood cell loss and BARC-4 bleeding occurs with increasing platelet reactivity in patients on antiplatelet therapy undergoing CABG. Our findings support current guidelines to determine time of surgery based on an objective measurement of platelet function (Platelet Inhibition and Bleeding in Patients Undergoing Emergent Cardiac Surgery; clinicaltrials.gov NCT01468597).
T
[3]. Furthermore, blood transfusion was associated with a dose-dependent increase in morbidity and mortality after CABG [4]. To avoid excessive bleeding, current guidelines recommend a standardized 5 days preoperative period of withdrawal of clopidogrel and ticagrelor and a 7 days withdrawal of prasugrel, unless there is high ischemic risk [5].
hree to 15% of patients presenting with acute coronary syndrome undergo coronary artery bypass grafting (CABG) during dual antiplatelet therapy (DAPT) with aspirin and a P2Y12 receptor inhibitor [1, 2]. Although preoperative P2Y12 receptor inhibitor therapy has been associated with a reduction in the risk of ischemic event occurrences, recent exposure has been associated with an increased relative risk of death and reoperation by about 50 and 200%, respectively. The latter observations are mainly attributed to excessive bleeding Accepted for publication May 2, 2016. Presented at the Poster Session of the American Heart Association Scientific Sessions, Chicago, IL, Nov 15–19, 2014. Address correspondence to Dr Gurbel, Inova Center for Thrombosis Research and Drug Development, Inova Heart and Vascular Institute, 3300 Gallows Rd, Falls Church, VA 22042; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Dr Mahla discloses a financial relationship with Astra Zeneca, CSL Behring, and Boehringer Ingelheim; Dr Gurbel with Daiichi Sankyo, Bayer, Medtronic, MedImmune, Duke Clinical Research Institute, Coramed, Janssen, National Institute of Health, AstraZeneca, Boehringer, Merck, CSL, New Haven Pharmaceuticals, and Haemonetics.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2016.05.003
2
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
The Timing Based on Platelet Function Strategy to Reduce Clopidogrel Associated Bleeding Related to CABG (TARGET-CABG) study prospectively demonstrated that a platelet function measurement-based strategy to time CABG in clopidogrel-treated patients was associated with the same amount of bleeding that occurred in clopidogrel-naive patients and with a 50% shorter preoperative waiting period than recommended in the guidelines [6]. In patients who require urgent CABG during DAPT, current guidelines recommend to time surgery based on tests of platelet function [5, 7]. However, the optimal platelet function assay and a validated cutoff of platelet reactivity predicting bleeding remain elusive [7]. Although the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction 38 (TRITON-TIMI) and Study of Platelet Inhibition and Patient Outcomes (PLATO) trial results suggested that CABG-related bleeding increased with greater P2Y12 receptor inhibition, studies evaluating the relationship between measured platelet reactivity and CABG-related bleeding in patients on aspirin and new P2Y12 receptor inhibitors are lacking [8, 9]. Thus, the aim of the current study was to evaluate whether or not on-treatment platelet reactivity as assessed by light transmittance aggregometry (LTA) was associated with on-pump CABG-related bleeding in patients recently exposed to DAPT. Furthermore, we also sought to assess the utility of other platelet function tests and the correlation between platelet function tests.
Ann Thorac Surg 2016;-:-–-
dose of 300 IU/kg unfractionated heparin to obtain an activated clotting time 400 seconds that was maintained during cardiopulmonary bypass (CPB) by supplemental administration. On completion of CPB, anticoagulation was reversed by protamine chloride in a 1:1 ratio; additional protamine was given as needed to achieve an activated clotting time <140 seconds. All patients received tranexamic acid. The perioperative transfusion trigger was set to a hematocrit of 20% on CPB and 25% thereafter, unless active bleeding or low cardiac output suggested a need to increase this level. Platelets were administered in abnormal postpump bleeding based on clinical judgment [10]. Thrombelastometry (ROTEM, Tem Innovations GmbH, Munich, Germany) was used to guide fresh frozen plasma, Prothromplex (Baxter Healthcare GmbH, Vienna Austria), and Haemocomplettan (CSL Behring GmbH, Vienna Austria) administration to target the normal range of clotting time and clot strength in tissue factor activated (EXTEM), heparinase-containing (HEPTEM) and fibrin-based (FIBTEM) thrombelasotmetry. As per institutional protocol rethoracotomy was performed in case of hemodynamic instability, cardiac tamponade, and chest tube drainage of 1,000 mL after excluding or correcting residual heparin or dilutional coagulopathy. Patients received 325 mg aspirin intravenously 6 hours postoperatively and 100 mg/day thereafter unless contraindicated by active bleeding. One-year mortality was assessed by telephone interview.
Platelet Function Testing Patients and Methods After institutional review board approval (Medical University of Graz, Austria; EK: 21 to 202 ex 09/10), all patients with recent exposure to DAPT undergoing urgent or emergent CABG with or without concomitant valve replacement were screened between October 2010 and September 2013 for eligibility for this prospective observational study. After obtaining written informed consent, patients were enrolled in the study if they received aspirin (100 to 500 mg/day) and a P2Y12 receptor inhibitor within 48 hours preoperatively. Patients on chronic dialysis, on concomitant oral anticoagulants and patients unable to consent were excluded. Patients undergoing off-pump CABG or concomitant valve replacement and patients without platelet function data were excluded from final analysis.
Blood samples were obtained by venipuncture within 2 to 4 hours preoperatively and platelet reactivity was assessed by the following: Chronolog 700 Lumi-Aggregometer (Chronolog Corp., Havertown, PA) using 5 mM adenosine phosphate (ADP) to stimulate platelet aggregation in platelet-rich plasma [11] Flow cytometric vasodilator-stimulated phosphoprotein (VASP)-phosphorylation assay (Biocytex, Marseille, France) [11] Multiplate analyzer (Roche Diagnostics GmbH, Vienna, Austria) using the ADP test [12] Innovance-PFA2Y (Siemens Healthcare, Marburg, Germany) using the P2Y cartridge in whole blood as described previously [13].
Patient Management
Study Endpoints
DAPT was discontinued once an indication for CABG was established. As per institutional protocol, glycoprotein IIb or IIIa inhibitors were stopped preoperatively for at least 4 hours and 8 hours in patients with normal renal function and a glomerular filtration rate 30 mL/min, respectively. The perioperative care of the patients was at the discretion of the attending physicians, who were blinded to platelet function data. Surgery was performed by the individual cardiac surgeon on duty. Anticoagulation was established by an initial loading
The primary endpoint was the calculated perioperative red blood cell (RBC) loss that has been used previously to assess bleeding in patients undergoing cardiac surgery and was calculated as follows: (blood volume preoperative hematocrit 0.91) – (blood volume hematocrit 0.91 on postoperative day 5) þ (mL of transfused RBCs 0.59). A factor of 0.91 was applied to correct hematocrit of peripheral blood sampling and the factor 0.59 accounts for the average hematocrit of RBC units [14].
Ann Thorac Surg 2016;-:-–-
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
3
The secondary endpoint was Bleeding Academic Research Consortium (BARC)-4 bleeding defined as the following: 48 hours of perioperative intracranial bleeding Reoperation after closure of sternotomy Transfusion of 5 units of packed RBCs within 48 hours 24-hour chest tube drainage 2000 mL [15]
Statistical Analysis As estimations of study effect were unavailable, a formal power calculation could not be performed. It was anticipated that 60 to 70 patients per year would be recruited. Therefore, with a study duration of 3 years, and a dropout rate of 10%, a total of about 175 patients was expected. Categorical data are presented as absolute and relative frequencies and continuous data are presented as mean SD or as median (interquartile range) as appropriate. Platelet reactivity as measured by LTA (5 mm ADP), VASP-platelet reactivity index (PRI) and Multiplate ADP test is presented in tertiles, closure time as measured by Innovance PFA2Y is considered as a binary variable (cutoff 300 seconds). For categorical variables, the chi-square test was used for comparison, whereas an analysis of variance (ANOVA) or Kruskal-Wallis test was used for continuous data. Post hoc t tests or Mann-Whitney U tests were performed as appropriate applying a Bonferroni correction. Spearman correlation coefficients were calculated to assess the association between LTA and the various platelet function tests. Changes in hemoglobin, fibrinogen and platelet count over time were assessed by a repeated measures ANOVA. To study the impact of different variables on RBC loss, linear regression analysis was performed. The following variables were considered: EuroSCORE [16], aspirin dose, CPB time, duration of surgery, use of glycoprotein IIb or IIIa inhibitors, preoperative platelet count, tertiles of platelet reactivity as assessed by LTA (5 mm ADP), Multiplate ADP test, and VASP-PRI and closure time measured by Innovance PFA2Y [7, 10]. Furthermore, a multivariable model containing EuroSCORE, aspirin dose, CPB time, and LTA-assessed tertile distribution of platelet reactivity was considered. The level of significance was set at 0.05. Statistical analysis was performed using SPSS Version 22 (IBM SPSS Statistics, Chicago, IL) and R version 3.1.1.
Results The study flow diagram is presented in Figure 1. One hundred forty-nine patients underwent on-pump CABG during DAPT with aspirin and clopidogrel (n ¼ 80), aspirin and prasugrel (n ¼ 28), or aspirin and ticagrelor (n ¼ 41). In 74% of patients, CABG was performed within 24 hours of the last dose and in 26% within 48 hours. Overall, preoperative platelet reactivity was low [median (IQR): LTA (5 mM ADP): 20 (9–28)%; VASP-PRI: 39 (15–73)%; Multiplate-ADP test: 16 (12–22) U*min; and
Fig 1. Study flow diagram. (CABG ¼ coronary artery bypass grafting.)
Innovance-PFA2Y 300 sec: 72%]. Correlation between different platelet function assays was variable (R ¼ 0.54 for 5 mM ADP-induced LTA versus VASP-PRI; R ¼ 0.45 for 5 mM ADP-induced LTA versus Multiplate ADP test; and R¼ –0.53 for 5 mm ADP-induced LTA versus Innovance PFA2Y; p < 0.001 for all). Baseline, procedural, and outcome characteristics of the patients in relation to LTA-assessed tertile distribution of platelet reactivity are given in Tables 1 and 2. Duration of surgery, CPB time, units of heparin, and protamine were similar across tertiles. In 5 patients who underwent rethoracotomy, a surgical cause for bleeding was identified. The 30-day mortality rate was 9%. Four patients died from cardiac and 9 patients died from noncardiac causes. None of the 131 patients available for long-term follow-up died during a median of 509 (414 to 762) days after hospital discharge. Tertile distribution of platelet reactivity as assessed by LTA and VASP-PRI demonstrated increased bleeding associated with decreased platelet reactivity (Table 3). There was a significant difference in calculated RBC loss, BARC-4 bleeding, 24-hour chest tube drainage, RBC transfusion during 48 and 120 hours, and platelet transfusion across LTA assessed tertiles. Bleeding variables were significantly higher in the first tertile, as compared to the third tertile. VASP-PRI assessed tertile distribution of platelet reactivity demonstrated significant differences in RBC and platelet transfusions between the groups. Transfusion requirements were significantly higher in the first tertile as compared to the third tertile.
4
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Ann Thorac Surg 2016;-:-–-
Table 1. Baseline Characteristics in Relation to Tertile Distribution of Platelet Reactivity as Assessed by Light Transmittance Aggregometry (% Aggregation) Variable Age, years Male Body mass index EuroSCORE Serum creatinine, mg/dL Hypertension Diabetes Acute coronary syndrome Ejection fraction, % Left main disease Aspirin dose 100 mg 300 mg Glycoprotein inhibitors
Total (n ¼ 149)
First Tertile (0–13.7%)
Second Tertile (13.7–25%)
Third Tertile (25–100%)
68.2 9.4 114 (76.5) 26.6 (24.4–29.4) 8.4 4.3 0.9 (0.8–1.2) 117 (78.5) 34 (22.8) 147 (98.7) 45.8 12.3 79 (53.0)
67.8 8.7 39 (78.0) 27.6 (25.6–29.8) 7.7 4.3 1.0 (0.8–1.2) 40 (80.0) 11 (22.0) 50 (100.0) 45.6 12.3 28 (56.0)
67.8 9.8 38 (73.1) 26.4 (24.3–28.4) 8.6 3.9 0.9 (0.8–1.2) 39 (75.0) 10 (19.2) 51 (98.1) 46.9 11.9 32 (61.5)
69.2 9.8 37 (78.7) 26.0 (24.0–29.1) 9.0 4.7 0.9 (0.8–1.2) 38 (80.9) 13 (27.7) 46 (97.9) 44.7 12.9 19 (40.4)
107 (71.8) 42 (28.2) 5 (3.4)
32 (64.0) 18 (36.0) 5 (10.0)
38 (73.1) 14 (26.9) 0 (0.0)
37 (78.7) 10 (21.3) 0 (0.0)
Values are mean SD, n (%), or median (interquartile range). EuroSCORE ¼ European system for cardiac operative risk evaluation;
Bleeding and transfusion in relation to closure time as assessed by Innovance PFA2Y is presented in Table 4. Compared to patients with a closure time of equal to or more than 300 seconds those with a closure time less than 300 seconds had a lower incidence of BARC-4 bleeding and lower transfusion rates of RBCs and platelets. Perioperative hemoglobin and fibrinogen levels were similar between the LTA-assessed tertiles at each time point and changed significantly over time (p < 0.001 for trend) (Fig 2). Median (IQR) platelet count decreased from 204 (178 to 240) to 141 (115 to 170) on ICU arrival and only slowly recovered on postoperative day 5 to 170 (134 to 214; p < 0.001 for trend [data not shown]). The association between aspirin dose, EuroSCORE, CPB time, duration of surgery, preoperative glycoprotein
SD ¼ standard deviation.
inhibitors, preoperative platelet count, platelet reactivity as determined by various platelet function tests, and calculated RBC loss as assessed by univariate regression analysis is presented in Table 5. Among the platelet function tests, LTA and VASP-PRI were significantly associated with RBC loss. In a multivariable regression model, aspirin dose, CPB time EuroSCORE, and tertile distribution of LTA-assessed platelet reactivity remained significantly associated with RBC loss (Table 6).
Comment To the best of our knowledge, this is the first prospective study evaluating the association between tertile distribution of preoperative on-treatment platelet reactivity
Table 2. Procedural and Outcome Characteristics in Relation to Tertile Distribution of Platelet Reactivity as Assessed by Light Transmittance Aggregometry (% Aggregation) Variable CPB time, min Duration of surgery, min Heparin, 1,000 units Protamin, 1,000 units Saphenous grafts Left internal mammary artery Intensive care unit stay, days Length of hospital stay, days Rethoracotomy 30-day mortality a
Total (n ¼ 149)
First Tertile (0–13.7%)
Second Tertile (13.7–25%)
Third Tertile (25–100%)
p Value
108.7 36.7 239.8 54.6 33.0 (26.0–41.0) 37.2 13.0 2.0 (2.0–3.0) 83 (55.7) 6.0 (4.0–11.0) 15.0 (12.0–22.0) 15 (10.1) 13 (9.0)
111.9 43.4 237.0 50.1 34.5 (26.5–40.0) 35.3 13.5 2.0 (2.0–3.0) 26 (52.0) 8.0 (5.0–14.8) 18.0 (14.0–28.8) 4 (8.0) 5 (10.2)
107.9 37.4 249.7 62.6 33.0 (26.0–41.0) 38.8 14.7 2.0 (2.0–3.0) 32 (61.5) 5.0 (3.0–8.2) 13.0 (11.0–19.2) 5 (9.6) 4 (8.0)
106.0 27.7 232.0 48.9 31.0 (27.5–40.0) 37.4 10.4 2.0 (2.0–3.0) 25 (53.2) 7.0 (4.0–10.5) 17.0 (14.0–22.0) 6 (12.8) 4 (8.7)
0.724 0.246 0.961 0.399 0.979 0.574 0.046a 0.004a 0.745 1.000
Significant difference between first and second tertiles (Bonferroni correction).
Values are mean SD, median (interquartile range), or n (%). CBP ¼ cardiopulmonary bypass;
SD ¼ standard deviation.
Ann Thorac Surg 2016;-:-–-
5
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Table 3. Bleeding and Transfusion in Relation to Tertile Distribution of Platelet Reactivity Total (n ¼ 149)
Variable LTA, 5 mm ADP induced, % aggregation Calculated red blood cell loss, mL BARC-4 bleeding 24h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0–120 h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h VASP-PRI, % Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0-120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h Multiplate ADP test, U*min Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0-120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h a
1,196 72 1,115 4.0 5.0 1.0 25 70
(937–1,572) (48.3) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (16.8) (47.0)
1,196 69 1,115 4.0 5.0 1.0 24 68
(937–1,572) (47.3) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (16.4) (46.6)
1,196 72 1,115 4.0 5.0 1.0 25 69
(937–1,572) (49.0) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (17.0) (46.9)
First Tertile
Second Tertile
Third Tertile
0–13.7%
13.7–25%
25–100%
1,449 31 1,295 5.0 6.0 1.5 10 26
1,107 (858–1,517) 24 (46.2) 1,030 (722–1,505) 4.0 (2.0–5.2) 5.0 (3.0–7.0) 1.0 (0.0–2.0) 8 (15.4) 24 (46.2) 26.8–65% 1,092 (971–1,455) 22 (45.8) 1,160 (871–1,506) 4.0 (2.0–6.0) 5.0 (2.8–7.0) 1.0 (0.8–2.0) 7 (14.6) 19 (39.6) 13–20 U*min 1,072 (890–1,565) 24 (42.9) 1,115 (730–1,480) 4.0 (1.8–6.0) 5.0 (2.0–7.2) 1.0 (0.0–2.0) 8 (14.3) 27 (48.2)
1,075 (811–1,269) 17 (36.2) 935 (674–1,332) 2.0 (0.5–5.0) 3.0 (1.5–6.0) 1.0 (0.0–1.0) 7 (14.9) 20 (42.6) 65–100% 1,085 (746–1,425) 18 (36.7) 965 (672–1,555) 3.0 (1.0–5.0) 4.0 (2.0–6.0) 1.0 (0.0–2.0) 6 (12.2) 23 (46.9) 20–83 U*min 1,175 (956–1,443) 17 (41.5) 920 (702–1,618) 3.0 (2.0–5.0) 4.0 (3.0–7.0) 1.0 (0.0–1.0) 6 (14.6) 16 (39.0)
(1,020–1,754) (62.0) (872–1,900) (3.2–7.0) (4.0–7.8) (1.0–2.0) (20.0) (52.0) 0–26.8% 1,333 (1,017–1,735) 29 (59.2) 1130 (700–1,695) 5.0 (3.0–7.0) 5.0 (3.0–8.0) 2.0 (1.0–2.0) 11 (22.4) 26 (53.1) 1–13 U*min 1,259 (1,014–1,682) 31 (62.0) 1,328 (846–1,619) 5.0 (2.2–7.0) 5.5 (3.0–7.0) 1.5 (1.0–2.0) 11 (22.0) 26 (52.0)
p Value
0.004a 0.037 0.035a 0.005a 0.014a 0.008a 0.762 0.644 0.057 0.081 0.522 0.022a 0.027a 0.041a 0.427 0.444 0.511 0.078 0.307 0.289 0.521 0.122 0.554 0.465
Significant difference between first and third tertiles (Bonferroni correction).
Values are median (interquartile range) or n (%). ADP ¼ adenosine phosphate; BARC ¼ Bleeding Academic Research Consortium; FFP ¼ fresh-frozen plasma; LTA ¼ light transmittance aggregometry; RBC ¼ red blood cells; VASP-PRI ¼ vasodilator-stimulated phosphoprotein platelet reactivity index.
and bleeding in patients undergoing on-pump CABG within 48 hours of last DAPT including the new P2Y12 receptor inhibitors. Calculated RBC loss, BARC-4 bleeding, 24-hour chest tube drainage, and RBC and
platelet transfusion decreased with increasing LTA assessed tertiles of platelet reactivity, suggesting less bleeding with higher on-treatment platelet reactivity. Similarly, transfusion of RBCs and platelets decreased
Table 4. Bleeding and Transfusion in Relation to Platelet Reactivity Measured by Innovance PFA2Y
Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0–120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h
<300 s (n ¼ 41)
Total (n ¼ 146)
Closure Time
1,196 70 1,115 4.0 5.0 1.0 25 67
(937–1,572) (47.9) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (17.1) (45.9)
1,070 14 1,005.0 3.0 4.0 1.0 3 19
(866–1,319) (34.1) (620–1,450) (2.0–5.0) (2.0–6.0) (0.0–1.0) (7.3) (46.3)
Values are median (interquartile range) or n (%). BARC ¼ Bleeding Academic Research Consortium;
FFP ¼ fresh-frozen plasma;
RBC ¼ red blood cells.
300 s (n ¼ 105) 1,225 56 1,170 4.0 5.0 1.0 22 48
(975–1,684) (53.3) (794–1,654) (2.0–7.0) (3.0–8.0) (1.0–2.0) (21.0) (45.7)
p Value 0.065 0.044 0.160 0.028 0.041 0.005 0.054 1.000
6
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Ann Thorac Surg 2016;-:-–-
Fig 2. (A) Perioperative change of hemoglobin (g/L) and (B) fibrinogen (mg/dL) according to light transmittance aggregometry (LTA)–assessed tertile of platelet reactivity presented as box plots. (Preop. ¼ 2 to 4 hours preoperatively; ICU ¼ intensive care unit; POD ¼ postoperative day.)
with increasing platelet reactivity as assessed by VASPPRI but not the Multiplate ADP test. The latter is most probably due to a very low median platelet inhibition of 20 U*min, which is very close to “low platelet reactivity” and the narrow interquartile range [17]. Moreover there was less BARC-4 bleeding and less RBC and platelet transfusion with a closure time of less than 300 seconds
as compared to a closure time equal to or more than 300 seconds. The available literature suggests that increased CABGrelated bleeding is associated with recent exposure to P2Y12 receptor inhibitors and more potent P2Y12 receptor inhibitors, albeit using variable bleeding definitions [3, 15]. In a nationwide registry including 2,244 patients,
Ann Thorac Surg 2016;-:-–-
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Table 5. Variables Associated With Calculated Red Blood Cell Loss in Univariate Linear Regression Analysis Variable
p Value
Aspirin dose EuroSCORE Cardiopulmonary bypass time Duration of surgery Glycoprotein inhibitors Preoperative platelet count LTA, 5 mm ADP-induced % aggregation (tertiles) VASP-PRI, % aggregation (tertiles) Multiplate ADP test (tertiles) Innovance PFA2Y (binary)
0.014 0.023 0.002 0.007 0.131 0.868 0.003 0.025 0.694 0.133
ADP ¼ adenosine diphosphate; EuroSCORE ¼ European system for cardiac operative risk evaluation; LTA ¼ light transmittance aggregometry; VASP-PRI ¼ vasodilator-stimulated phosphoprotein platelet reactivity index.
Hansson and colleagues [18] recently demonstrated the benefits of the reversibility of ticagrelor in comparison to clopidogrel. While there was a continuous decrease in major CABG-related bleeding with increasing days off clopidogrel, there was no difference in major bleeding beyond a 3- to 5-day preoperative waiting period in ticagrelor-treated patients. Small studies in clopidogreltreated patients demonstrated a graded association between preoperatively measured platelet reactivity levels and increased bleeding [19, 20]. Moreover, an individualized strategy for preoperative clopidogrel cessation based on platelet function testing reduced both preoperative waiting and surgery-related bleeding in on- and off-pump CABG [6, 21]. However, validated platelet reactivity cutoffs associated with bleeding and the optimal platelet function assay to predict bleeding remain elusive so far and studies evaluating platelet reactivity and bleeding in patients on new P2Y12 receptor inhibitors are lacking [7, 18]. Furthermore, there is a
Table 6. Multivariable Linear Regression Model for Calculated Red Blood Cell Loss 95% Confidence Interval
b
SE
LB
UB
(Constant) 802.1 162.4 480.9 1123.3 Aspirin dose 300 mg 206.3 94.5 19.5 393.1 Cardiopulmonary 3.5 1.1 1.35 5.7 bypass time (min) LTA, % aggregation –245.2 96.5 436.0 54.4 (second tertile) LTA, % aggregation –346.5 99.1 542.4 150.6 (third tertile) EuroSCORE 23.0 9.7 3.8 42.2
p Value <0.001 0.031 0.002 0.012 0.001 0.019
Adjusted R2 ¼ 0.189. EuroSCORE ¼ European system for cardiac operative risk evaluation; LB ¼ lower bound; LTA ¼ light transmittance aggregometry; SE ¼ standard error; UB ¼ upper bound.
7
wide variability in transfusion rates in patients undergoing CABG and it is reported that a 20% incidence of preoperative anemia may increase transfusion requirements irrespective of surgery-related bleeding [14]. To reliably assess a potential association between measured platelet reactivity and bleeding in moderate to high risk patients presenting for CABG on DAPT, we therefore used different platelet function tests and evaluated their relation to calculated RBC loss. The latter involved preoperative and postoperative hematocrit values as well as units of transfused RBCs and hidden blood loss from hematoma [14]. There was a parallel perioperative change in hemoglobin among the 3 LTAassessed tertiles of platelet inhibition in our study suggesting a consistent transfusion practice. Moreover, pre- and immediate postoperative fibrinogen levels were similar across tertiles and well above the recommended target plasma cutoffs to treat surgical bleeding [22]. Mannacio and colleagues [21] scheduled off-pump CABG after individualized waiting in patients on DAPT resulted in a clotting time 106 seconds, suggesting complete recovery of platelet function. TARGET-CABG suggested that high platelet reactivity as assessed by TEG platelet mapping served as a surrogate for adequate surgical hemostasis in on-pump CABG [6]. In patients undergoing CABG or combined procedures during DAPT Ranucci and colleagues [20] identified 31 U*min and CPB time as independent determinants of bleeding. Kwak and colleagues [19] demonstrated 70% platelet inhibition as the optimal cutoff for predicting transfusion after off-pump CABG. However, on-pump CABG, though employing antifibrinolytics and targeting high normal postoperative fibrinogen levels, might require higher platelet reactivity to obviate bleeding and transfusion. An impact of high dose aspirin on CABGassociated bleeding has been suggested previously and is conceivable due to antiplatelet effects that are dosedependent [23–25]. The clinical implications of our findings are that in patients undergoing urgent surgery with recent exposure to DAPT every additional hour on pump increases RBC loss by about 220 mL and that increasing tertiles of platelet reactivity decreased RBC loss by 245 and 345 mL, respectively. There were no prespecified platelet transfusion triggers. Open-label therapy with a potent P2Y12 inhibitor may have biased excessive platelet transfusions in the ticagrelor- and prasugrel-treated patients. Surgical bleeding is likely more heterogeneous in etiology than bleeding in the PCI-treated patient. Correcting for the multitude of nonhematologic factors that influence surgical bleeding is a formidable endeavor despite extensive statistical analyses. The EuroSCORE may overestimate calculated cardiac risk but EuroSCORE II was not yet available at study start. In conclusion, although our study in patients undergoing urgent CABG with recent exposure to DAPT can neither provide a bleeding cutoff nor define the optimal platelet function assay the results strongly suggest a graded decrease in RBC loss and BARC-4 bleeding with increasing ranges of platelet reactivity. The results
8
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
support current guidelines that suggest to time surgery based on platelet function monitoring rather than using fixed drug-related preoperative waiting periods and to delay urgent surgery for at least 1 to 2 days. Dr Mahla received grants from CSL Behring and Novo Nordisk; Dr Gurbel from the National Institutes of Health, Daiichi Sankyo, Harvard Research Institute, New Haven Pharmaceuticals, Merck, Coramed, Haemonetics, and the Duke Clinical Research Institute. Dr Gurbel holds patents in the area of personalized antiplatelet therapy and interventional cardiology. The authors would like to acknowledge Simone Tischler, Tobias Niedrist, and Raphael Neururer for their technical assistance.
References 1. Held C, Asenblad N, Bassand JP, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes undergoing coronary artery bypass surgery: results from the PLATO (platelet inhibition and patient outcomes) trial. J Am Coll Cardiol 2011;57:672–84. 2. Smith PK, Goodnough LT, Levy JH, et al. Mortality benefit with prasugrel in the TRITON-TIMI 38 coronary artery bypass grafting cohort: risk-adjusted retrospective data analysis. J Am Coll Cardiol 2012;60:388–96. 3. Siller-Matula JM, Petre A, Delle-Karth G, et al. Impact of preoperative use of P2Y12 receptor inhibitors on clinical outcomes in cardiac and non-cardiac surgery: a systematic review and meta-analysis. Eur Heart J Acute Cardiovasc Care 2015 May 5. [Epub ahead of print]. 4. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA 2010;304:1559–67. 5. Windecker S, Kolh P, Alfonso F, et al. 2014 ESC/EACTS guidelines on myocardial revascularization: the task force on myocardial revascularization of the European Society of Cardiology (ESC) and the European Association for CardioThoracic Surgery (EACTS). Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J 2014;35: 2541–619. 6. Mahla E, Suarez TA, Bliden KP, et al. Platelet function measurement-based strategy to reduce bleeding and waiting time in clopidogrel-treated patients undergoing coronary artery bypass graft surgery: the timing based on platelet function strategy to reduce clopidogrel-associated bleeding related to CABG (Target-CABG) study. Circ Cardiovasc Interv 2012;5:261–9. 7. Ferraris VA, Saha SP, Oestreich JH, et al. 2012 update to the Society of Thoracic Surgeons guideline on use of antiplatelet drugs in patients having cardiac and noncardiac operations. Ann Thorac Surg 2012;94:1761–81. 8. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2009;361:1045–57. 9. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2007;357:2001–15.
Ann Thorac Surg 2016;-:-–-
10. Ferraris VA, Brown JR, Despotis GJ, et al. 2011 Update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists Blood Conservation Clinical Practice Guidelines. Ann Thorac Surg 2011;91:944–82. 11. Gurbel PA, Bliden KP, Butler K, et al. Randomized doubleblind assessment of the ONSET and OFFSet of the antiplatelet Effects of Ticagrelor versus clopidogrel in patients with stable coronary artery disease: The ONSET/OFFSET Study. Circulation 2009;120:2577–85. 12. Sibbing D, Schulz S, Braun S, et al. Antiplatelet effects of clopidogrel and bleeding in patients undergoing coronary stent placement. J Thromb Haemost 2010;8:250–6. 13. Tsantes A, Ikonomidis I, Papadakis I, et al. Evaluation of the role of the new innovance PFA P2Y test cartridge in detection of clopidogrel resistance. Platelets 2012;23:481–9. 14. Gombotz H, Rehak PH, Shander A, Hofmann A. Blood use in elective surgery: the Austrian Benchmark Study. Transfusion 2007;47:1468–80. 15. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation 2011;123:2736–47. 16. Nashef SA, Roques F, Michel P, et al. European system for cardiac operative risk evaluation (Euroscore). Eur J Cardiothorac Surg 1999;16:9–13. 17. Tantry US, Bonello L, Aradi D, et al. Consensus and update on the definition of on-treatment platelet reactivity to adenosine diphosphate associated with ischemia and bleeding. J Am Coll Cardiol 2013;62:2261–73. 18. Hansson EC, Jideus L, Aberg B, et al. Coronary artery bypass grafting-related bleeding complications in patients treated with ticagrelor or clopidogrel: a nationwide study. Eur Heart J 2016;37:189–97. 19. Kwak YL, Kim JC, Choi YS, et al. Clopidogrel responsiveness regardless of the discontinuation date predicts increased blood loss and transfusion requirement after off-pump coronary artery bypass graft surgery. J Am Coll Cardiol 2010;56: 1994–2002. 20. Ranucci M, Baryshnikova E, Soro G, et al. Multiple electrode whole-blood aggregometry and bleeding in cardiac surgery patients receiving thienopyridines. Ann Thorac Surg 2011;91: 123–9. 21. Mannacio V, Meier P, Antignano A, et al. Individualized strategy for clopidogrel suspension in patients undergoing offpump coronary surgery for acute coronary syndrome: a casecontrol study. J Thorac Cardiovasc Surg 2014;148:1299–306. 22. Levy JH, Welsby I, Goodnough LT. Fibrinogen as a therapeutic target for bleeding: a review of critical levels and replacement therapy. Transfusion 2014;54:1389–405. 23. Sun JC, Whitlock R, Cheng J, et al. The effect of pre-operative aspirin on bleeding, transfusion, myocardial infarction, and mortality in coronary artery bypass surgery: a systematic review of randomized and observational studies. Eur Heart J 2008;29:1057–71. 24. Undas A, Brummel-Ziedins K, Mann KG. Why does aspirin decrease the risk of venous thromboembolism? On old and novel antithrombotic effects of acetyl salicylic acid. J Thromb Haemost 2014;12:1776–87. 25. Gurbel PA, Bliden KP, DiChiara J, et al. Evaluation of doserelated effects of aspirin on platelet function: results from the ASpirin-Induced Platelet EffeCT (ASPECT) study. Circulation 2007;115:3156–64.
Background. Up to 15% of patients require coronary artery bypass grafting (CABG) during dual antiplatelet therapy. Available evidence suggests an association between platelet reactivity and CABG-related bleeding. However, platelet reactivity cutoffs for bleeding remain elusive. We sought to explore the association between platelet reactivity and bleeding. Methods. Patients on aspirin and a P2Y12 receptor inhibitor within 48 hours before isolated CABG (n [ 149) were enrolled in this prospective study. Blood was drawn 2 to 4 hours preoperatively and platelet reactivity assessed by light transmittance aggregometry (LTA), vasodilatorstimulated phosphoprotein (VASP) assay, Multiplate analyzer and Innovance PFA2Y. The primary endpoint was calculated red blood cell loss computed as follows: (blood volume 3 preoperative hematocrit 3 0.91) – (blood volume 3 hematocrit 3 0.91 on postoperative day 5) D (mL of transfused red blood cells 3 0.59). Results. Preoperative platelet reactivity was low [median (interquartile range): LTA: 20 (9–28)%; VASPPRI: 39 (15–73)%; Multiplate adenosine phosphate test: 16 (12–22) U*min]. Innovance PFA2Y ‡300 seconds,
72%. Median (IQR) red blood cell loss in patients in first the LTA tertile was 1,449 (1,020 to 1,754) mL compared with 1,107 (858 to 1,512) mL and 1,075 (811 to 1,269) mL in those in the second and third tertiles, respectively (p < 0.004). Bleeding Academic Research Consortium (BARC)-4 bleeding differed between tertiles (62% versus 46% versus 36%; p ¼ 0.037). In a multivariable linear regression model, aspirin dose ‡300 mg, cardiopulmonary bypass time, EuroSCORE, and tertile distribution of platelet reactivity were significantly associated with red blood cell loss. Conclusions. A gradual decrease in red blood cell loss and BARC-4 bleeding occurs with increasing platelet reactivity in patients on antiplatelet therapy undergoing CABG. Our findings support current guidelines to determine time of surgery based on an objective measurement of platelet function (Platelet Inhibition and Bleeding in Patients Undergoing Emergent Cardiac Surgery; clinicaltrials.gov NCT01468597).
T
[3]. Furthermore, blood transfusion was associated with a dose-dependent increase in morbidity and mortality after CABG [4]. To avoid excessive bleeding, current guidelines recommend a standardized 5 days preoperative period of withdrawal of clopidogrel and ticagrelor and a 7 days withdrawal of prasugrel, unless there is high ischemic risk [5].
hree to 15% of patients presenting with acute coronary syndrome undergo coronary artery bypass grafting (CABG) during dual antiplatelet therapy (DAPT) with aspirin and a P2Y12 receptor inhibitor [1, 2]. Although preoperative P2Y12 receptor inhibitor therapy has been associated with a reduction in the risk of ischemic event occurrences, recent exposure has been associated with an increased relative risk of death and reoperation by about 50 and 200%, respectively. The latter observations are mainly attributed to excessive bleeding Accepted for publication May 2, 2016. Presented at the Poster Session of the American Heart Association Scientific Sessions, Chicago, IL, Nov 15–19, 2014. Address correspondence to Dr Gurbel, Inova Center for Thrombosis Research and Drug Development, Inova Heart and Vascular Institute, 3300 Gallows Rd, Falls Church, VA 22042; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Dr Mahla discloses a financial relationship with Astra Zeneca, CSL Behring, and Boehringer Ingelheim; Dr Gurbel with Daiichi Sankyo, Bayer, Medtronic, MedImmune, Duke Clinical Research Institute, Coramed, Janssen, National Institute of Health, AstraZeneca, Boehringer, Merck, CSL, New Haven Pharmaceuticals, and Haemonetics.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2016.05.003
2
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
The Timing Based on Platelet Function Strategy to Reduce Clopidogrel Associated Bleeding Related to CABG (TARGET-CABG) study prospectively demonstrated that a platelet function measurement-based strategy to time CABG in clopidogrel-treated patients was associated with the same amount of bleeding that occurred in clopidogrel-naive patients and with a 50% shorter preoperative waiting period than recommended in the guidelines [6]. In patients who require urgent CABG during DAPT, current guidelines recommend to time surgery based on tests of platelet function [5, 7]. However, the optimal platelet function assay and a validated cutoff of platelet reactivity predicting bleeding remain elusive [7]. Although the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction 38 (TRITON-TIMI) and Study of Platelet Inhibition and Patient Outcomes (PLATO) trial results suggested that CABG-related bleeding increased with greater P2Y12 receptor inhibition, studies evaluating the relationship between measured platelet reactivity and CABG-related bleeding in patients on aspirin and new P2Y12 receptor inhibitors are lacking [8, 9]. Thus, the aim of the current study was to evaluate whether or not on-treatment platelet reactivity as assessed by light transmittance aggregometry (LTA) was associated with on-pump CABG-related bleeding in patients recently exposed to DAPT. Furthermore, we also sought to assess the utility of other platelet function tests and the correlation between platelet function tests.
Ann Thorac Surg 2016;-:-–-
dose of 300 IU/kg unfractionated heparin to obtain an activated clotting time 400 seconds that was maintained during cardiopulmonary bypass (CPB) by supplemental administration. On completion of CPB, anticoagulation was reversed by protamine chloride in a 1:1 ratio; additional protamine was given as needed to achieve an activated clotting time <140 seconds. All patients received tranexamic acid. The perioperative transfusion trigger was set to a hematocrit of 20% on CPB and 25% thereafter, unless active bleeding or low cardiac output suggested a need to increase this level. Platelets were administered in abnormal postpump bleeding based on clinical judgment [10]. Thrombelastometry (ROTEM, Tem Innovations GmbH, Munich, Germany) was used to guide fresh frozen plasma, Prothromplex (Baxter Healthcare GmbH, Vienna Austria), and Haemocomplettan (CSL Behring GmbH, Vienna Austria) administration to target the normal range of clotting time and clot strength in tissue factor activated (EXTEM), heparinase-containing (HEPTEM) and fibrin-based (FIBTEM) thrombelasotmetry. As per institutional protocol rethoracotomy was performed in case of hemodynamic instability, cardiac tamponade, and chest tube drainage of 1,000 mL after excluding or correcting residual heparin or dilutional coagulopathy. Patients received 325 mg aspirin intravenously 6 hours postoperatively and 100 mg/day thereafter unless contraindicated by active bleeding. One-year mortality was assessed by telephone interview.
Platelet Function Testing Patients and Methods After institutional review board approval (Medical University of Graz, Austria; EK: 21 to 202 ex 09/10), all patients with recent exposure to DAPT undergoing urgent or emergent CABG with or without concomitant valve replacement were screened between October 2010 and September 2013 for eligibility for this prospective observational study. After obtaining written informed consent, patients were enrolled in the study if they received aspirin (100 to 500 mg/day) and a P2Y12 receptor inhibitor within 48 hours preoperatively. Patients on chronic dialysis, on concomitant oral anticoagulants and patients unable to consent were excluded. Patients undergoing off-pump CABG or concomitant valve replacement and patients without platelet function data were excluded from final analysis.
Blood samples were obtained by venipuncture within 2 to 4 hours preoperatively and platelet reactivity was assessed by the following: Chronolog 700 Lumi-Aggregometer (Chronolog Corp., Havertown, PA) using 5 mM adenosine phosphate (ADP) to stimulate platelet aggregation in platelet-rich plasma [11] Flow cytometric vasodilator-stimulated phosphoprotein (VASP)-phosphorylation assay (Biocytex, Marseille, France) [11] Multiplate analyzer (Roche Diagnostics GmbH, Vienna, Austria) using the ADP test [12] Innovance-PFA2Y (Siemens Healthcare, Marburg, Germany) using the P2Y cartridge in whole blood as described previously [13].
Patient Management
Study Endpoints
DAPT was discontinued once an indication for CABG was established. As per institutional protocol, glycoprotein IIb or IIIa inhibitors were stopped preoperatively for at least 4 hours and 8 hours in patients with normal renal function and a glomerular filtration rate 30 mL/min, respectively. The perioperative care of the patients was at the discretion of the attending physicians, who were blinded to platelet function data. Surgery was performed by the individual cardiac surgeon on duty. Anticoagulation was established by an initial loading
The primary endpoint was the calculated perioperative red blood cell (RBC) loss that has been used previously to assess bleeding in patients undergoing cardiac surgery and was calculated as follows: (blood volume preoperative hematocrit 0.91) – (blood volume hematocrit 0.91 on postoperative day 5) þ (mL of transfused RBCs 0.59). A factor of 0.91 was applied to correct hematocrit of peripheral blood sampling and the factor 0.59 accounts for the average hematocrit of RBC units [14].
Ann Thorac Surg 2016;-:-–-
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
3
The secondary endpoint was Bleeding Academic Research Consortium (BARC)-4 bleeding defined as the following: 48 hours of perioperative intracranial bleeding Reoperation after closure of sternotomy Transfusion of 5 units of packed RBCs within 48 hours 24-hour chest tube drainage 2000 mL [15]
Statistical Analysis As estimations of study effect were unavailable, a formal power calculation could not be performed. It was anticipated that 60 to 70 patients per year would be recruited. Therefore, with a study duration of 3 years, and a dropout rate of 10%, a total of about 175 patients was expected. Categorical data are presented as absolute and relative frequencies and continuous data are presented as mean SD or as median (interquartile range) as appropriate. Platelet reactivity as measured by LTA (5 mm ADP), VASP-platelet reactivity index (PRI) and Multiplate ADP test is presented in tertiles, closure time as measured by Innovance PFA2Y is considered as a binary variable (cutoff 300 seconds). For categorical variables, the chi-square test was used for comparison, whereas an analysis of variance (ANOVA) or Kruskal-Wallis test was used for continuous data. Post hoc t tests or Mann-Whitney U tests were performed as appropriate applying a Bonferroni correction. Spearman correlation coefficients were calculated to assess the association between LTA and the various platelet function tests. Changes in hemoglobin, fibrinogen and platelet count over time were assessed by a repeated measures ANOVA. To study the impact of different variables on RBC loss, linear regression analysis was performed. The following variables were considered: EuroSCORE [16], aspirin dose, CPB time, duration of surgery, use of glycoprotein IIb or IIIa inhibitors, preoperative platelet count, tertiles of platelet reactivity as assessed by LTA (5 mm ADP), Multiplate ADP test, and VASP-PRI and closure time measured by Innovance PFA2Y [7, 10]. Furthermore, a multivariable model containing EuroSCORE, aspirin dose, CPB time, and LTA-assessed tertile distribution of platelet reactivity was considered. The level of significance was set at 0.05. Statistical analysis was performed using SPSS Version 22 (IBM SPSS Statistics, Chicago, IL) and R version 3.1.1.
Results The study flow diagram is presented in Figure 1. One hundred forty-nine patients underwent on-pump CABG during DAPT with aspirin and clopidogrel (n ¼ 80), aspirin and prasugrel (n ¼ 28), or aspirin and ticagrelor (n ¼ 41). In 74% of patients, CABG was performed within 24 hours of the last dose and in 26% within 48 hours. Overall, preoperative platelet reactivity was low [median (IQR): LTA (5 mM ADP): 20 (9–28)%; VASP-PRI: 39 (15–73)%; Multiplate-ADP test: 16 (12–22) U*min; and
Fig 1. Study flow diagram. (CABG ¼ coronary artery bypass grafting.)
Innovance-PFA2Y 300 sec: 72%]. Correlation between different platelet function assays was variable (R ¼ 0.54 for 5 mM ADP-induced LTA versus VASP-PRI; R ¼ 0.45 for 5 mM ADP-induced LTA versus Multiplate ADP test; and R¼ –0.53 for 5 mm ADP-induced LTA versus Innovance PFA2Y; p < 0.001 for all). Baseline, procedural, and outcome characteristics of the patients in relation to LTA-assessed tertile distribution of platelet reactivity are given in Tables 1 and 2. Duration of surgery, CPB time, units of heparin, and protamine were similar across tertiles. In 5 patients who underwent rethoracotomy, a surgical cause for bleeding was identified. The 30-day mortality rate was 9%. Four patients died from cardiac and 9 patients died from noncardiac causes. None of the 131 patients available for long-term follow-up died during a median of 509 (414 to 762) days after hospital discharge. Tertile distribution of platelet reactivity as assessed by LTA and VASP-PRI demonstrated increased bleeding associated with decreased platelet reactivity (Table 3). There was a significant difference in calculated RBC loss, BARC-4 bleeding, 24-hour chest tube drainage, RBC transfusion during 48 and 120 hours, and platelet transfusion across LTA assessed tertiles. Bleeding variables were significantly higher in the first tertile, as compared to the third tertile. VASP-PRI assessed tertile distribution of platelet reactivity demonstrated significant differences in RBC and platelet transfusions between the groups. Transfusion requirements were significantly higher in the first tertile as compared to the third tertile.
4
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Ann Thorac Surg 2016;-:-–-
Table 1. Baseline Characteristics in Relation to Tertile Distribution of Platelet Reactivity as Assessed by Light Transmittance Aggregometry (% Aggregation) Variable Age, years Male Body mass index EuroSCORE Serum creatinine, mg/dL Hypertension Diabetes Acute coronary syndrome Ejection fraction, % Left main disease Aspirin dose 100 mg 300 mg Glycoprotein inhibitors
Total (n ¼ 149)
First Tertile (0–13.7%)
Second Tertile (13.7–25%)
Third Tertile (25–100%)
68.2 9.4 114 (76.5) 26.6 (24.4–29.4) 8.4 4.3 0.9 (0.8–1.2) 117 (78.5) 34 (22.8) 147 (98.7) 45.8 12.3 79 (53.0)
67.8 8.7 39 (78.0) 27.6 (25.6–29.8) 7.7 4.3 1.0 (0.8–1.2) 40 (80.0) 11 (22.0) 50 (100.0) 45.6 12.3 28 (56.0)
67.8 9.8 38 (73.1) 26.4 (24.3–28.4) 8.6 3.9 0.9 (0.8–1.2) 39 (75.0) 10 (19.2) 51 (98.1) 46.9 11.9 32 (61.5)
69.2 9.8 37 (78.7) 26.0 (24.0–29.1) 9.0 4.7 0.9 (0.8–1.2) 38 (80.9) 13 (27.7) 46 (97.9) 44.7 12.9 19 (40.4)
107 (71.8) 42 (28.2) 5 (3.4)
32 (64.0) 18 (36.0) 5 (10.0)
38 (73.1) 14 (26.9) 0 (0.0)
37 (78.7) 10 (21.3) 0 (0.0)
Values are mean SD, n (%), or median (interquartile range). EuroSCORE ¼ European system for cardiac operative risk evaluation;
Bleeding and transfusion in relation to closure time as assessed by Innovance PFA2Y is presented in Table 4. Compared to patients with a closure time of equal to or more than 300 seconds those with a closure time less than 300 seconds had a lower incidence of BARC-4 bleeding and lower transfusion rates of RBCs and platelets. Perioperative hemoglobin and fibrinogen levels were similar between the LTA-assessed tertiles at each time point and changed significantly over time (p < 0.001 for trend) (Fig 2). Median (IQR) platelet count decreased from 204 (178 to 240) to 141 (115 to 170) on ICU arrival and only slowly recovered on postoperative day 5 to 170 (134 to 214; p < 0.001 for trend [data not shown]). The association between aspirin dose, EuroSCORE, CPB time, duration of surgery, preoperative glycoprotein
SD ¼ standard deviation.
inhibitors, preoperative platelet count, platelet reactivity as determined by various platelet function tests, and calculated RBC loss as assessed by univariate regression analysis is presented in Table 5. Among the platelet function tests, LTA and VASP-PRI were significantly associated with RBC loss. In a multivariable regression model, aspirin dose, CPB time EuroSCORE, and tertile distribution of LTA-assessed platelet reactivity remained significantly associated with RBC loss (Table 6).
Comment To the best of our knowledge, this is the first prospective study evaluating the association between tertile distribution of preoperative on-treatment platelet reactivity
Table 2. Procedural and Outcome Characteristics in Relation to Tertile Distribution of Platelet Reactivity as Assessed by Light Transmittance Aggregometry (% Aggregation) Variable CPB time, min Duration of surgery, min Heparin, 1,000 units Protamin, 1,000 units Saphenous grafts Left internal mammary artery Intensive care unit stay, days Length of hospital stay, days Rethoracotomy 30-day mortality a
Total (n ¼ 149)
First Tertile (0–13.7%)
Second Tertile (13.7–25%)
Third Tertile (25–100%)
p Value
108.7 36.7 239.8 54.6 33.0 (26.0–41.0) 37.2 13.0 2.0 (2.0–3.0) 83 (55.7) 6.0 (4.0–11.0) 15.0 (12.0–22.0) 15 (10.1) 13 (9.0)
111.9 43.4 237.0 50.1 34.5 (26.5–40.0) 35.3 13.5 2.0 (2.0–3.0) 26 (52.0) 8.0 (5.0–14.8) 18.0 (14.0–28.8) 4 (8.0) 5 (10.2)
107.9 37.4 249.7 62.6 33.0 (26.0–41.0) 38.8 14.7 2.0 (2.0–3.0) 32 (61.5) 5.0 (3.0–8.2) 13.0 (11.0–19.2) 5 (9.6) 4 (8.0)
106.0 27.7 232.0 48.9 31.0 (27.5–40.0) 37.4 10.4 2.0 (2.0–3.0) 25 (53.2) 7.0 (4.0–10.5) 17.0 (14.0–22.0) 6 (12.8) 4 (8.7)
0.724 0.246 0.961 0.399 0.979 0.574 0.046a 0.004a 0.745 1.000
Significant difference between first and second tertiles (Bonferroni correction).
Values are mean SD, median (interquartile range), or n (%). CBP ¼ cardiopulmonary bypass;
SD ¼ standard deviation.
Ann Thorac Surg 2016;-:-–-
5
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Table 3. Bleeding and Transfusion in Relation to Tertile Distribution of Platelet Reactivity Total (n ¼ 149)
Variable LTA, 5 mm ADP induced, % aggregation Calculated red blood cell loss, mL BARC-4 bleeding 24h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0–120 h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h VASP-PRI, % Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0-120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h Multiplate ADP test, U*min Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0-120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h a
1,196 72 1,115 4.0 5.0 1.0 25 70
(937–1,572) (48.3) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (16.8) (47.0)
1,196 69 1,115 4.0 5.0 1.0 24 68
(937–1,572) (47.3) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (16.4) (46.6)
1,196 72 1,115 4.0 5.0 1.0 25 69
(937–1,572) (49.0) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (17.0) (46.9)
First Tertile
Second Tertile
Third Tertile
0–13.7%
13.7–25%
25–100%
1,449 31 1,295 5.0 6.0 1.5 10 26
1,107 (858–1,517) 24 (46.2) 1,030 (722–1,505) 4.0 (2.0–5.2) 5.0 (3.0–7.0) 1.0 (0.0–2.0) 8 (15.4) 24 (46.2) 26.8–65% 1,092 (971–1,455) 22 (45.8) 1,160 (871–1,506) 4.0 (2.0–6.0) 5.0 (2.8–7.0) 1.0 (0.8–2.0) 7 (14.6) 19 (39.6) 13–20 U*min 1,072 (890–1,565) 24 (42.9) 1,115 (730–1,480) 4.0 (1.8–6.0) 5.0 (2.0–7.2) 1.0 (0.0–2.0) 8 (14.3) 27 (48.2)
1,075 (811–1,269) 17 (36.2) 935 (674–1,332) 2.0 (0.5–5.0) 3.0 (1.5–6.0) 1.0 (0.0–1.0) 7 (14.9) 20 (42.6) 65–100% 1,085 (746–1,425) 18 (36.7) 965 (672–1,555) 3.0 (1.0–5.0) 4.0 (2.0–6.0) 1.0 (0.0–2.0) 6 (12.2) 23 (46.9) 20–83 U*min 1,175 (956–1,443) 17 (41.5) 920 (702–1,618) 3.0 (2.0–5.0) 4.0 (3.0–7.0) 1.0 (0.0–1.0) 6 (14.6) 16 (39.0)
(1,020–1,754) (62.0) (872–1,900) (3.2–7.0) (4.0–7.8) (1.0–2.0) (20.0) (52.0) 0–26.8% 1,333 (1,017–1,735) 29 (59.2) 1130 (700–1,695) 5.0 (3.0–7.0) 5.0 (3.0–8.0) 2.0 (1.0–2.0) 11 (22.4) 26 (53.1) 1–13 U*min 1,259 (1,014–1,682) 31 (62.0) 1,328 (846–1,619) 5.0 (2.2–7.0) 5.5 (3.0–7.0) 1.5 (1.0–2.0) 11 (22.0) 26 (52.0)
p Value
0.004a 0.037 0.035a 0.005a 0.014a 0.008a 0.762 0.644 0.057 0.081 0.522 0.022a 0.027a 0.041a 0.427 0.444 0.511 0.078 0.307 0.289 0.521 0.122 0.554 0.465
Significant difference between first and third tertiles (Bonferroni correction).
Values are median (interquartile range) or n (%). ADP ¼ adenosine phosphate; BARC ¼ Bleeding Academic Research Consortium; FFP ¼ fresh-frozen plasma; LTA ¼ light transmittance aggregometry; RBC ¼ red blood cells; VASP-PRI ¼ vasodilator-stimulated phosphoprotein platelet reactivity index.
and bleeding in patients undergoing on-pump CABG within 48 hours of last DAPT including the new P2Y12 receptor inhibitors. Calculated RBC loss, BARC-4 bleeding, 24-hour chest tube drainage, and RBC and
platelet transfusion decreased with increasing LTA assessed tertiles of platelet reactivity, suggesting less bleeding with higher on-treatment platelet reactivity. Similarly, transfusion of RBCs and platelets decreased
Table 4. Bleeding and Transfusion in Relation to Platelet Reactivity Measured by Innovance PFA2Y
Calculated red blood cell loss, mL BARC-4 bleeding 24 h chest tube drainage, mL RBC transfusion (units) 0–48 h RBC transfusion (units) 0–120h Platelets (units) 0–48 h FFP 0–48 h Fibrinogen 0–48 h
<300 s (n ¼ 41)
Total (n ¼ 146)
Closure Time
1,196 70 1,115 4.0 5.0 1.0 25 67
(937–1,572) (47.9) (725–1,618) (2.0–6.0) (3.0–7.0) (0.0–2.0) (17.1) (45.9)
1,070 14 1,005.0 3.0 4.0 1.0 3 19
(866–1,319) (34.1) (620–1,450) (2.0–5.0) (2.0–6.0) (0.0–1.0) (7.3) (46.3)
Values are median (interquartile range) or n (%). BARC ¼ Bleeding Academic Research Consortium;
FFP ¼ fresh-frozen plasma;
RBC ¼ red blood cells.
300 s (n ¼ 105) 1,225 56 1,170 4.0 5.0 1.0 22 48
(975–1,684) (53.3) (794–1,654) (2.0–7.0) (3.0–8.0) (1.0–2.0) (21.0) (45.7)
p Value 0.065 0.044 0.160 0.028 0.041 0.005 0.054 1.000
6
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Ann Thorac Surg 2016;-:-–-
Fig 2. (A) Perioperative change of hemoglobin (g/L) and (B) fibrinogen (mg/dL) according to light transmittance aggregometry (LTA)–assessed tertile of platelet reactivity presented as box plots. (Preop. ¼ 2 to 4 hours preoperatively; ICU ¼ intensive care unit; POD ¼ postoperative day.)
with increasing platelet reactivity as assessed by VASPPRI but not the Multiplate ADP test. The latter is most probably due to a very low median platelet inhibition of 20 U*min, which is very close to “low platelet reactivity” and the narrow interquartile range [17]. Moreover there was less BARC-4 bleeding and less RBC and platelet transfusion with a closure time of less than 300 seconds
as compared to a closure time equal to or more than 300 seconds. The available literature suggests that increased CABGrelated bleeding is associated with recent exposure to P2Y12 receptor inhibitors and more potent P2Y12 receptor inhibitors, albeit using variable bleeding definitions [3, 15]. In a nationwide registry including 2,244 patients,
Ann Thorac Surg 2016;-:-–-
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
Table 5. Variables Associated With Calculated Red Blood Cell Loss in Univariate Linear Regression Analysis Variable
p Value
Aspirin dose EuroSCORE Cardiopulmonary bypass time Duration of surgery Glycoprotein inhibitors Preoperative platelet count LTA, 5 mm ADP-induced % aggregation (tertiles) VASP-PRI, % aggregation (tertiles) Multiplate ADP test (tertiles) Innovance PFA2Y (binary)
0.014 0.023 0.002 0.007 0.131 0.868 0.003 0.025 0.694 0.133
ADP ¼ adenosine diphosphate; EuroSCORE ¼ European system for cardiac operative risk evaluation; LTA ¼ light transmittance aggregometry; VASP-PRI ¼ vasodilator-stimulated phosphoprotein platelet reactivity index.
Hansson and colleagues [18] recently demonstrated the benefits of the reversibility of ticagrelor in comparison to clopidogrel. While there was a continuous decrease in major CABG-related bleeding with increasing days off clopidogrel, there was no difference in major bleeding beyond a 3- to 5-day preoperative waiting period in ticagrelor-treated patients. Small studies in clopidogreltreated patients demonstrated a graded association between preoperatively measured platelet reactivity levels and increased bleeding [19, 20]. Moreover, an individualized strategy for preoperative clopidogrel cessation based on platelet function testing reduced both preoperative waiting and surgery-related bleeding in on- and off-pump CABG [6, 21]. However, validated platelet reactivity cutoffs associated with bleeding and the optimal platelet function assay to predict bleeding remain elusive so far and studies evaluating platelet reactivity and bleeding in patients on new P2Y12 receptor inhibitors are lacking [7, 18]. Furthermore, there is a
Table 6. Multivariable Linear Regression Model for Calculated Red Blood Cell Loss 95% Confidence Interval
b
SE
LB
UB
(Constant) 802.1 162.4 480.9 1123.3 Aspirin dose 300 mg 206.3 94.5 19.5 393.1 Cardiopulmonary 3.5 1.1 1.35 5.7 bypass time (min) LTA, % aggregation –245.2 96.5 436.0 54.4 (second tertile) LTA, % aggregation –346.5 99.1 542.4 150.6 (third tertile) EuroSCORE 23.0 9.7 3.8 42.2
p Value <0.001 0.031 0.002 0.012 0.001 0.019
Adjusted R2 ¼ 0.189. EuroSCORE ¼ European system for cardiac operative risk evaluation; LB ¼ lower bound; LTA ¼ light transmittance aggregometry; SE ¼ standard error; UB ¼ upper bound.
7
wide variability in transfusion rates in patients undergoing CABG and it is reported that a 20% incidence of preoperative anemia may increase transfusion requirements irrespective of surgery-related bleeding [14]. To reliably assess a potential association between measured platelet reactivity and bleeding in moderate to high risk patients presenting for CABG on DAPT, we therefore used different platelet function tests and evaluated their relation to calculated RBC loss. The latter involved preoperative and postoperative hematocrit values as well as units of transfused RBCs and hidden blood loss from hematoma [14]. There was a parallel perioperative change in hemoglobin among the 3 LTAassessed tertiles of platelet inhibition in our study suggesting a consistent transfusion practice. Moreover, pre- and immediate postoperative fibrinogen levels were similar across tertiles and well above the recommended target plasma cutoffs to treat surgical bleeding [22]. Mannacio and colleagues [21] scheduled off-pump CABG after individualized waiting in patients on DAPT resulted in a clotting time 106 seconds, suggesting complete recovery of platelet function. TARGET-CABG suggested that high platelet reactivity as assessed by TEG platelet mapping served as a surrogate for adequate surgical hemostasis in on-pump CABG [6]. In patients undergoing CABG or combined procedures during DAPT Ranucci and colleagues [20] identified 31 U*min and CPB time as independent determinants of bleeding. Kwak and colleagues [19] demonstrated 70% platelet inhibition as the optimal cutoff for predicting transfusion after off-pump CABG. However, on-pump CABG, though employing antifibrinolytics and targeting high normal postoperative fibrinogen levels, might require higher platelet reactivity to obviate bleeding and transfusion. An impact of high dose aspirin on CABGassociated bleeding has been suggested previously and is conceivable due to antiplatelet effects that are dosedependent [23–25]. The clinical implications of our findings are that in patients undergoing urgent surgery with recent exposure to DAPT every additional hour on pump increases RBC loss by about 220 mL and that increasing tertiles of platelet reactivity decreased RBC loss by 245 and 345 mL, respectively. There were no prespecified platelet transfusion triggers. Open-label therapy with a potent P2Y12 inhibitor may have biased excessive platelet transfusions in the ticagrelor- and prasugrel-treated patients. Surgical bleeding is likely more heterogeneous in etiology than bleeding in the PCI-treated patient. Correcting for the multitude of nonhematologic factors that influence surgical bleeding is a formidable endeavor despite extensive statistical analyses. The EuroSCORE may overestimate calculated cardiac risk but EuroSCORE II was not yet available at study start. In conclusion, although our study in patients undergoing urgent CABG with recent exposure to DAPT can neither provide a bleeding cutoff nor define the optimal platelet function assay the results strongly suggest a graded decrease in RBC loss and BARC-4 bleeding with increasing ranges of platelet reactivity. The results
8
MAHLA ET AL PLATELET REACTIVITY AND CABG-RELATED BLEEDING
support current guidelines that suggest to time surgery based on platelet function monitoring rather than using fixed drug-related preoperative waiting periods and to delay urgent surgery for at least 1 to 2 days. Dr Mahla received grants from CSL Behring and Novo Nordisk; Dr Gurbel from the National Institutes of Health, Daiichi Sankyo, Harvard Research Institute, New Haven Pharmaceuticals, Merck, Coramed, Haemonetics, and the Duke Clinical Research Institute. Dr Gurbel holds patents in the area of personalized antiplatelet therapy and interventional cardiology. The authors would like to acknowledge Simone Tischler, Tobias Niedrist, and Raphael Neururer for their technical assistance.
References 1. Held C, Asenblad N, Bassand JP, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes undergoing coronary artery bypass surgery: results from the PLATO (platelet inhibition and patient outcomes) trial. J Am Coll Cardiol 2011;57:672–84. 2. Smith PK, Goodnough LT, Levy JH, et al. Mortality benefit with prasugrel in the TRITON-TIMI 38 coronary artery bypass grafting cohort: risk-adjusted retrospective data analysis. J Am Coll Cardiol 2012;60:388–96. 3. Siller-Matula JM, Petre A, Delle-Karth G, et al. Impact of preoperative use of P2Y12 receptor inhibitors on clinical outcomes in cardiac and non-cardiac surgery: a systematic review and meta-analysis. Eur Heart J Acute Cardiovasc Care 2015 May 5. [Epub ahead of print]. 4. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA 2010;304:1559–67. 5. Windecker S, Kolh P, Alfonso F, et al. 2014 ESC/EACTS guidelines on myocardial revascularization: the task force on myocardial revascularization of the European Society of Cardiology (ESC) and the European Association for CardioThoracic Surgery (EACTS). Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J 2014;35: 2541–619. 6. Mahla E, Suarez TA, Bliden KP, et al. Platelet function measurement-based strategy to reduce bleeding and waiting time in clopidogrel-treated patients undergoing coronary artery bypass graft surgery: the timing based on platelet function strategy to reduce clopidogrel-associated bleeding related to CABG (Target-CABG) study. Circ Cardiovasc Interv 2012;5:261–9. 7. Ferraris VA, Saha SP, Oestreich JH, et al. 2012 update to the Society of Thoracic Surgeons guideline on use of antiplatelet drugs in patients having cardiac and noncardiac operations. Ann Thorac Surg 2012;94:1761–81. 8. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2009;361:1045–57. 9. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2007;357:2001–15.
Ann Thorac Surg 2016;-:-–-
10. Ferraris VA, Brown JR, Despotis GJ, et al. 2011 Update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists Blood Conservation Clinical Practice Guidelines. Ann Thorac Surg 2011;91:944–82. 11. Gurbel PA, Bliden KP, Butler K, et al. Randomized doubleblind assessment of the ONSET and OFFSet of the antiplatelet Effects of Ticagrelor versus clopidogrel in patients with stable coronary artery disease: The ONSET/OFFSET Study. Circulation 2009;120:2577–85. 12. Sibbing D, Schulz S, Braun S, et al. Antiplatelet effects of clopidogrel and bleeding in patients undergoing coronary stent placement. J Thromb Haemost 2010;8:250–6. 13. Tsantes A, Ikonomidis I, Papadakis I, et al. Evaluation of the role of the new innovance PFA P2Y test cartridge in detection of clopidogrel resistance. Platelets 2012;23:481–9. 14. Gombotz H, Rehak PH, Shander A, Hofmann A. Blood use in elective surgery: the Austrian Benchmark Study. Transfusion 2007;47:1468–80. 15. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation 2011;123:2736–47. 16. Nashef SA, Roques F, Michel P, et al. European system for cardiac operative risk evaluation (Euroscore). Eur J Cardiothorac Surg 1999;16:9–13. 17. Tantry US, Bonello L, Aradi D, et al. Consensus and update on the definition of on-treatment platelet reactivity to adenosine diphosphate associated with ischemia and bleeding. J Am Coll Cardiol 2013;62:2261–73. 18. Hansson EC, Jideus L, Aberg B, et al. Coronary artery bypass grafting-related bleeding complications in patients treated with ticagrelor or clopidogrel: a nationwide study. Eur Heart J 2016;37:189–97. 19. Kwak YL, Kim JC, Choi YS, et al. Clopidogrel responsiveness regardless of the discontinuation date predicts increased blood loss and transfusion requirement after off-pump coronary artery bypass graft surgery. J Am Coll Cardiol 2010;56: 1994–2002. 20. Ranucci M, Baryshnikova E, Soro G, et al. Multiple electrode whole-blood aggregometry and bleeding in cardiac surgery patients receiving thienopyridines. Ann Thorac Surg 2011;91: 123–9. 21. Mannacio V, Meier P, Antignano A, et al. Individualized strategy for clopidogrel suspension in patients undergoing offpump coronary surgery for acute coronary syndrome: a casecontrol study. J Thorac Cardiovasc Surg 2014;148:1299–306. 22. Levy JH, Welsby I, Goodnough LT. Fibrinogen as a therapeutic target for bleeding: a review of critical levels and replacement therapy. Transfusion 2014;54:1389–405. 23. Sun JC, Whitlock R, Cheng J, et al. The effect of pre-operative aspirin on bleeding, transfusion, myocardial infarction, and mortality in coronary artery bypass surgery: a systematic review of randomized and observational studies. Eur Heart J 2008;29:1057–71. 24. Undas A, Brummel-Ziedins K, Mann KG. Why does aspirin decrease the risk of venous thromboembolism? On old and novel antithrombotic effects of acetyl salicylic acid. J Thromb Haemost 2014;12:1776–87. 25. Gurbel PA, Bliden KP, DiChiara J, et al. Evaluation of doserelated effects of aspirin on platelet function: results from the ASpirin-Induced Platelet EffeCT (ASPECT) study. Circulation 2007;115:3156–64.