Combined clopidogrel and aspirin treatment up to surgery increases the risk of postoperative myocardial infarction, blood loss and reoperation for bleeding in patients undergoing coronary artery bypass grafting

Abstract

OBJECTIVES

Recent guidelines suggest that patients undergoing coronary artery bypass grafting (CABG) should discontinue clopidogrel and aspirin (ASA) 5 and 2–10 days, respectively, before surgery to reduce postoperative bleeding and its complications. The aim of our study was to evaluate the relationship between the timing of discontinuing clopidogrel + ASA and early clinical outcomes in patients undergoing CABG.

METHODS

Four thousand three hundred and thirty consecutive patients underwent isolated CABG from April 2004 to February 2009. Of these, 926 patients received double antiplatelet therapy in the 14 days prior to surgery. Patients were stratified into three groups: clopidogrel + ASA within 5 and 2 days, respectively, before surgery (Group A, n = 287); clopidogrel within 5 days + ASA stopped 2–10 days before surgery or clopidogrel stopped 5 days + ASA within 2 days of surgery (Group B, n = 308) and clopidogrel + ASA discontinued >5 and 10 days, respectively, before surgery (control group, n = 331).

RESULTS

Overall mortality was 0.8%. The incidence of postoperative myocardial infarction (MI) was 5.2, 1 and 1.8% in Groups A, B and control, respectively (P = 0.004). Reoperation for bleeding occurred in 4.5, 2.9 and 1.2% (P = 0.04) and total chest drainage was 761 ± 473, 720 ± 421 and 687 ± 302 ml in Groups A, B and control, respectively (P = 0.06). Multivariable analysis revealed that Group A was an independent predictor of postoperative MI (P = 0.02), reoperation for bleeding (P = 0.02), blood transfusions (P = 0.003) and blood losses (P = 0.015).

CONCLUSIONS

Clopidogrel in combination with ASA up to the time of surgery is associated with an increased risk of postoperative MI, blood loss and reoperation for bleeding in patients undergoing CABG.

INTRODUCTION

Combined clopidogrel and aspirin (ASA) treatment has been shown to reduce mortality and morbidity in the setting of acute coronary syndrome (ACS) and percutaneous coronary intervention, making double antiplatelet therapy a class I recommendation and Grade A evidence in the American College of Cardiology and America Heart Association guideline [1]. Consequently, most patients with ACS receive these drugs. There are still significant controversies regarding the preoperative use of double antiplatet treatment in patients undergoing coronary artery bypass grafting (CABG). Several retrospective studies have shown that clopidogrel in association with ASA given <5 days before surgery is associated with more blood products, re-explorations for bleeding and an increase in chest drain blood loss [2–6]. Nevertheless, randomized trials have suggested that patients undergoing CABG after clopidogrel and ASA benefit from adverse anti-ischaemic effects without an increase in life-threatening bleeding. The authors of the Clopidogrel in Unstable angina to prevent Recurrent ischaemic Events (CURE) and Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) found, respectively, 1.7 and 4.6% risk reduction in cardiovascular events associated with 2.1 and 0.6% increased risk in major bleeding among CABG patients [7, 8]. However, subanalyses of these studies revealed that patients undergoing CABG within 5 days of receiving clopidogrel had, respectively, 5.2 and 5.5% increased risk for major bleeding and, interestingly, 3.1 and 5.7% increased risk for composite ischaemic events compared with patients who received clopidogrel >5 days before surgery. Recent guidelines suggest that patients undergoing CABG should discontinue clopidogrel and ASA 5 and 2–10 days, respectively, before surgery to reduce postoperative bleeding and its complications [9]. Despite these recommendations, 10–30% of patients who require urgent CABG are still operated on under the effect of these drugs [3, 10]. Therefore, the aim of our study was to evaluate the relationship between the timing of discontinuing clopidogrel in combination with ASA and early clinical outcomes in patients undergoing CABG.

MATERIALS AND METHODS

Patient selection, data collection and definitions

The study was approved by the clinical audit committee of the University Hospitals Bristol National Health Service Foundation Trust to meet ethical and legal requirements, and individual consent was waived. This was a retrospective, observational, cohort study of prospectively collected data from consecutive patients who underwent isolated CABG at the Bristol Heart Institute between April 2004 and February 2009. The data collection form is entered in a database (Patient Analysis and Tracking System; Dendrite Clinical Systems, London, UK) and includes five sections that are filled in consecutively by anaesthetists, surgeons, intensive care unit, high-dependency unit and ward nurses. The base sample contained detailed clinical information on ∼4330 patients. Of these, 3061 patients were excluded from the study because they were not on double antiplatelet treatment. The remaining patients (1269, 29%) underwent CABG surgery under the effect of double antiplatelet treatment.

Further exclusion criteria were patients who had unknown preoperative use or timing of ASA and clopidogrel (N = 24), unknown bleeding data (N = 208) and those with preoperative critical state (N = 57), defined as any one or more of the following: ventricular tachycardia or fibrillation or aborted sudden death, preoperative cardiac massage, preoperative ventilation before arrival in the anaesthetic room, preoperative inotropic support, intraaortic balloon counterpulsation or preoperative acute renal failure (anuria or oliguria <10 ml/h) or emergency operation (N = 54), defined as an operation carried out on referral before the beginning of the next working day). The final sample size was 926 patients who received double antiplatelet therapy in the last 14 days prior to surgery. Patients were stratified into three groups: clopidogrel + ASA within 5 and 2 days, respectively, before surgery (Group A, n = 287); clopidogrel within 5 days before surgery + ASA stopped 2–10 days before surgery or clopidogrel stopped 5 days prior to surgery + ASA within 2 days of surgery (Group B, n = 308) and clopidogrel + ASA discontinued >5 and 10 days, respectively, before surgery (control group, n = 331, Fig. 1). In-hospital mortality was defined as any death occurring within 30 days of operations. A diagnosis of postoperative myocardial infarction (MI) was based on the presence of Q-waves >0.04 ms and/or a reduction in R-waves >25% in at least two contiguous leads on electrocardiogram (ECG). Postoperative renal dysfunction was defined as a serum creatinine level >200 μmol/l. A diagnosis of stroke was made if there was evidence of new neurological deficit with morphological substrate confirmed by computer tomography or nuclear magnetic resonance imaging. Postoperative blood loss was defined as total chest tube drainage and indications for chest re-exploration were blood loss >500 ml over the first hour, >300 ml for two consecutive hours or >200 ml for three consecutive hours. Unstable angina was defined as rest angina requiring iv nitrates until arrival in the anaesthetic room. Finally, priority of surgery was defined as urgent (medical factors require the patient to stay in hospital waiting for an operation) and elective (the clinical status of the patient allows discharge from the hospital with readmission for surgery at a later date). The decision to perform off-pump or on-pump CABG technique was based on individual surgeon preference.

Figure 1:

Study profile.

Anaesthetic, surgical technique and postoperative management

Anaesthetic and surgical techniques were standardized for all patients and have been reported previously. Briefly, for patients undergoing on-pump CABG, cardiopulmonary bypass was instituted with the use of ascending aortic cannulation and two-stage venous cannulation of the right atrium. The membrane oxygenator was primed with 1000 ml of Hartmanns crystalloid, 500 ml gelofusine, 0.5 g/kg mannitol, 7 ml of 10% calcium gluconate and 6000 IU heparin. Alpha-stat pH management was used, and the systemic temperature was kept between 34 and 36°C, respectively. Myocardial protection was achieved with intermittent hyperkalaemic warm blood cardioplegia. For off-pump CABG, the Bristol technique was used to expose the coronaries and provides stabilization to undertake the anastomosis [11]. At the end of surgery, patients were transferred to the intensive care unit and managed according to the unit protocol. In the case of bleeding >150 ml for longer than consecutive 2 h, derangements of the coagulation profile were treated with a diagnosis-directed therapy. Elevation of activate clotting time >130 s was treated with additional 25 mg dose of protamine. Values of prothrombin time, activated partial thromboplastin time and international normalized ratio of >1.5 times control were corrected with fresh frozen plasma. A platelet count <80 000/µl was an indicator for platelet transfusion. A haematocrit <23% was the threshold for transfusion of red blood cell. The blood was all leucofiltrated. According to unit policy, patients received ASA 300 mg suppository per rectum within 6 h after surgery and then from the day after both 75 mg ASA and 75 mg clopidogrel per day for 1 year. In case of bleedings, antiplatelet treatment was delayed to the day after.

Statistical analysis

Continuous data were expressed as mean ± standard deviation (SD), and categorical data as percentages. The Kolmogorov–Smirnov test was used to check for normality of data before further analysis. Differences among groups were compared with the χ² test or Fisher's exact test for categorical variables and analysis of variance test with a post hoc Bonferroni test or Kruskal–Wallis test and Mann–Whitney U-tests with Bonferroni correction as appropriate for continuous variables. Multivariable logistic and linear regression analyses were performed to assess the effect of combined clopidogrel and ASA therapy on postoperative MI, re-exploration, blood transfusions and bleeding after adjusting for potentially confounding variables listed in Table 1. Clinically relevant variables with a P-value <0.2 on univariate analysis were incorporated in the multivariable models. Results are reported as percentages and odds ratios (ORs) with 95% confidence interval (CI). All reported P-values are two-sided, and P-values of <0.05 were considered to indicate statistical significance. All statistical analysis was performed with SPSS 15.0 (SPSS Inc., Chicago, IL, USA).

Table 1:

Preoperative baseline characteristics

Variable	Group A (n = 287)	Group B (n = 308)	Control group (n = 331)	P-value
Age (years ± SD)	65.5 ± 9	66 ± 9.9	65.5 ± 10.9	0.77
Female, n (%)	56 (19.5)	52 (16.9)	63 (19)	0.67
Diabetes, n (%)	49 (17.1)	54 (17.5)	72 (21.8)	0.25
Hypertension, n (%)	207 (72.1)	220 (71.4)	250 (75.5)	0.46
Chronic pulmonary disease, n (%)	27 (9.4)	31 (10.1)	37 (11.2)	0.76
Extracardiac arteriopathy, n (%)	20 (7)	30 (9.7)	35 (10.6)	0.28
Previous cardiac operations, n (%)	4 (1.4)	10 (3.2)	12 (3.6)	0.21
Preoperative EF (%), n (%)
Good	202 (70.4)	224 (72.7)	238 (71.9)	0.81
Fair	74 (25.8)	68 (22.1)	78 (23.6)
Poor	11 (3.8)	16 (5.2)	15 (4.5)
NYHA class III–IV, n (%)	73 (25.4)	70 (22.7)	92 (27.8)	0.34
CCS class 3–4	155 (54)*	162 (52.6)*	130 (39.3)	<0.0001
Recent ACS	186 (64.8)*§	167 (54.2)*	108 (32.6)	<0.0001
Unstable angina, n (%)	62 (21.6)*	67 (21.7)*	30 (9.1)	<0.0001
MI within 30 days, n (%)	152 (53)*§	116 (37.7)*	60 (18.1)	<0.0001
MI between 31–90 days, n (%)	25 (8.7)	45 (14.6)	46 (13.9)	0.06
MI >90 days, n (%)	30 (10.5)*	54 (17.5)*	128 (37.7)	<0.0001
Preoperative haemoglobin (mg/dl, SD)	13.3 ± 1.55*	13.5 ± 1.5	13.8 ± 1.4	<0.0001
Stop ASA (days ± SD)	0.9 ± 0.2*§	2.1 ± 1.5*	6.8 ± 2.4	<0.0001
Stop clopidogrel (days ± SD)	2.5 ± 1.6*§	6.6 ± 3.6*	8.4 ± 2.1	<0.0001
Heparin IV until operation	29 (10.1)	29 (9.4)	27 (8.2)	0.69
Aprotinin, n (%)	35 (12.2)*§	19 (6.2)	15 (4.5)	0.001
Tranexanic acid, n (%)	219 (76.3)	247 (80.2)	249 (75.2)	0.3
CAD, mean (SD)	2.6 ± 0.66	2.66 ± 0.58	2.63 ± 0.58	0.2
LMS >50%, n (%)	85 (29.6)*	93 (30.1)*	68 (20.5)	0.008
Urgent surgery, n (%)	243 (84.7)*§	233 (75.6)*	126 (38.1)	<0.0001
No. of grafts (mean, SD)	2.74 ± 0.77	2.72 ± 0.76	2.7 ± 0.78	0.72
Off-pump surgery, n (%)	177 (65.6)	210 (68.2)	217 (65.6)	0.25

EF: ejection fraction; CCS: Canadian cardiovascular society; ACS: acute coronary syndrome; MI: myocardial infarction; CAD: coronary artery disease; LMS: left main stenosis.

*P < 0.01 vs control.

^§ P < 0.01 vs Group B.

RESULTS

Preoperative baseline characteristics are outlined in Table 1. There was no difference in age, sex, cardiac risk factors, chronic pulmonary disease, extracardiac arteriopathy as well as left ventricular function. However, patients in Groups A and B had a higher incidence of left main disease, unstable angina and preoperative MI within 30 days, advanced Canadian cardiovascular society (CCS) class and urgent surgery as well as the higher intraoperative use of aprotinin than the control group. No difference was found among groups in the use of off-pump procedure. Postoperative clinical outcomes are shown in Table 2. Overall mortality was 0.8% with no difference among the three groups. The incidence of postoperative MI was 5.2, 1 and 1.8% in Groups A, B and control, respectively (P = 0.003). No difference was found in the incidence of postoperative renal dysfunction, atrial fibrillation and stroke. Reoperation for bleeding occurred in 4.5, 2.9 and 1.2% (P = 0.04) and total chest drainage was 761 ± 473, 720 ± 421 and 687 ± 302 ml in Groups A, B and control, respectively (P = 0.066). The mean number of red blood cells and platelets transfusion as well as the incidence of blood products transfused was higher in Group A (Table 2).

Table 2:

Early clinical outcomes

	Group A (N = 287)	Group B (N = 308)	Control group (N = 331)	P-value
Mortality, % (n)	1 (0.3)	0.6 (2)	1.2 (4)	0.45
Postoperative MI, % (n)	5.2 (15)*§	1 (3)	1.8 (6)	0.004
Stroke, % (n)	0.3 (1)	0.3 (1)	0.3 (1)	1
Re-exploration, % (n)	4.9 (14)*	2.9 (9)	1.2 (4)	0.04
PRD, % (n)	8.5 (24)	8.9 (27)	7.3 (24)	0.74
POAF, % (n)	24 (69)	23.4 (72)	28.1 (93)	0.33
Bleeding (ml ± SD)	761 ± 473	720 ± 421	687 ± 302	0.06
Transfusions (Unit)
Red blood cells	0.8 ± 1.6*§	0.5 ± 2.2	0.3 ± 0.9	0.001
Fresh frozen plasma	0.12 ± 0.67	0.9 ± 0.5	0.05 ± 0.02	0.33
Platelets	0.2 ± 0.6*§	0.1 ± 0.4	0.03 ± 0.2	<0.0001
Blood product exposure
Red blood cells	25.4 (73)*	17.5 (54)	15.1 (50)	0.003
Fresh frozen plasma	3.5 (10)	3.6 (11)	1.8 (6)	0.33
Platelets	15 (43)*§	7.1 (22)*	2.7 (9)	<0.0001

MI: myocardial infarction; PRD: postoperative renal dysfunction; POAF: postoperative atrial fibrillation.

*P < 0.01 vs control.

^§ P < 0.01 vs group B.

Postoperative myocardial infarction

Predictors of postoperative MI on univariate analysis were: clopidogrel + ASA within 5 and 2 days, respectively, before surgery (Group A, OR 3.26, 95% CI 1.43–7.4, P = 0.005), clopidogrel within 5 days before surgery + ASA stopped 2–10 days before surgery or clopidogrel stopped 5 days prior to surgery + ASA within 2 days of surgery (Group B, OR 0.4, 95% CI 0.14–1.2, P = 0.09), ejection fraction (EF) < 50% (OR 2, 95% CI 0.8–4.5, P = 0.12), hypertension (OR 0.5, 95% CI 0.2–1.2, P = 0.1), MI within 30 days before surgery (OR 1.9, 95% CI 0.8–4.2, P = 0.136) and number of diseased coronary arteries (OR 0.67, 95% CI 0.37–1.2, P = 0.175). However, on multivariable analysis, clopidogrel + ASA within 5 and 2 days, respectively, before surgery was the only independent risk factor for postoperative MI (OR 2.8, 95% CI 1.2–6.7, P = 0.02). Figure 2 shows that the risk of MI increased with decreasing the time from the last dose of clopidogrel and ASA.

Figure 2:

Probability of postoperative myocardial infarction and re-exploration for bleeding according to the time from last clopidogrel + ASA dose administration to the time of surgery.

Re-exploration for bleeding and blood loss

Predictors of re-exploration for bleeding on univariate analysis were EF < 30% (OR 2.9, 95% CI 0.8–10, P = 0.1), clopidogrel + ASA within 5 and 2 days, respectively, before surgery (Group A, OR 2.3, 95% CI 1.05–5, P = 0.036), NYHA class III–IV (OR 0.4, 95% CI 0.1–1.3, P = 0.011) and preoperative haemoglobin (OR 1.22, 95% CI 0.9–1.6, P = 0.15). Multivariable analysis showed that poor EF (OR 4.4, 95% CI 1.12–16.4, P = 0.026) and clopidogrel + ASA within 5 and 2 days, respectively, before surgery (OR 2.6, 95% CI 1.14–5.5, P = 0.023) were independent risk factors for re-exploration.

On univariate analysis, predictors of blood loss were clopidogrel + ASA within 5 and 2 days, respectively, before surgery (P = 0.04), age (P = 0.039), male gender (P < 0.0001), extracardiac arteriopathy (P = 0.008), redo operation (P = 0.12), unstable angina (P = 0.1), advanced NYHA class (P = 0.003), diabetes (P = 0.01), hypertension (P = 0.04), preoperative haemoglobin (P = 0.007), aprotinin (P = 0.04), tranexanic acid (P = 0.2), left main stenosis (P = 0.15) and off-pump surgery (P = 0.2). Multivariable linear regression identified that the use of clopidogrel and ASA within 5 and 2 days, respectively, before surgery (β = 0.08, P = 0.015) was an independent risk factor for bleeding. In addition, sex male (β = 0.09, P = 0.01), unstable angina (β = 0.1, P = 0.003), NYHA class III–IV (β = 0.72, P = 0.03), aprotinin (β = −0.16, P < 0.0001) and tranexanic acid (β = −0.13, P < 0.0001) were independent predictors for blood loss. Figures 2 and 3 show that the risk of re-exploration for bleeding and blood loss increased with decreasing the time from the last dose of clopidogrel and ASA.

Figure 3:

Predicted chest tube output according to the time from last clopidogrel + ASA dose administration to the time of surgery.

Blood transfusions

Predictors of univariate analysis were gender (OR 0.35, 95% CI 0.24–0.5, P < 0.0001), unstable angina (OR 2.7, 95% CI 1.8–2.9, P < 0.0001), chronic pulmonary disease (OR 2.4, 95% CI 1.5–2.8), EF < 50% (OR 2, 95% CI 1.4–2.7, P < 0.0001), preoperative haemoglobin (OR 0.55, 95% CI 0.5–0.6, P < 0.0001), clopidogrel + ASA within 5 and 2 days, respectively, before surgery (Group A, OR 1.8, 95% CI 1.3–2.5, P = 0.001), heparin IV (OR 1.8, 95%CI 1.3–2.6), off-pump procedures (OR 0.6, 95% CI 0.4–0.8, P = 0.003), advanced NYHA class (OR 1.6, 95% CI 1.1–2.3, P = 0.007), advanced CCS class (OR 1.5, 95% CI 1.1–2.1, P = 0.01), age (OR 1.05, 95% CI 1.02–1.07, P = 0.001), number of diseased coronary arteries (OR 1.5, 95% CI 1.02–2, P = 0.01) and urgent surgery (OR 1.4, 95% CI 1–2, P = 0.056).

On multivariable analysis, clopidogrel + ASA within 5 and 2 days, respectively, before surgery were an independent risk factor for blood transfusions (OR 1.9, 95% CI 1.2–2.9, P = 0.003). Other predictors were heparin IV (OR 2.7, 95% CI 1.5–4.7, P = 0.001), off-pump procedures (OR 0.45, 95% CI 0.3–0.8, P = 0.001), gender (OR 0.5, 95% CI 0.3–0.8, P = 0.003), pulmonary disease (OR 1.9, 95% CI 1.1–3.1, P = 0.03), EF < 50% (OR 1.8, 95% CI 1.2–2.8, P = 0.006), preoperative haemoglobin (OR 0.7, 95% CI 0.6–0.8, P < 0.0001) and advanced age (OR 1.04, 95% CI 1.02–1.06, P = 0.001).

DISCUSSION

Our study demonstrates that the use of clopidogrel in combination with ASA up to the time of surgery is associated with an increased risk of postoperative MI and reoperation for bleeding in patients undergoing CABG. Moreover, patients who were still under the effect of dual antiplatelet therapy had a higher blood loss as well as higher number and rate of blood transfusions. Finally, the probability of postoperative MI, re-exploration for bleeding and blood loss increased with decreasing time from the last clopidogrel and ASA dose.

Clopidogrel in addition to ASA has been demonstrated to reduce the risk of cardiovascular events in terms of death, MI, recurrent angina and stroke in the setting of ACS [12–14]. However, the use of double antiplatelet treatment in patients undergoing CABG is still controversial. Most observational studies and meta-analyses have shown that clopidogrel taken within 5–7 days of CABG is associated with an increased risk of bleeding and re-exploration, haemorrhage-related complications, increased rate of transfusion and greater consumption of health care resources [2–6, 15, 16]. Therefore, the recent guidelines on antiplatelet and anticoagulation management in cardiac surgery proposed by the European Association for Cardio-Thoracic Surgery recommend withdrawing clopidogrel and ASA 5 and 2–10 days, respectively, before surgery to reduce postoperative blood loss, risk of re-exploration and blood product usage; these benefits however, may be at the expense of a 1% increase in the risk of cardiovascular adverse events while awaiting surgery [9].

Nevertheless, several randomized trials have reported different results in terms of significant major bleedings [7, 13, 14]. In the CURE trial, Fox et al. [7], in a subanalysis of patients undergoing CABG, found clopidogrel to be associated with a 1.7% absolute risk reduction in the composite events of death, MI and stroke and a 1.4% absolute risk of life-threatening bleeding within 7 days after CABG. Among patients undergoing surgery enrolled for the CLARITY–TIMI 28 trial (clopidogrel as adjunctive reperfusion therapy–thrombolysis in MI), there was no difference in major or minor bleeding complications in patients who received clopidogrel within 5 days of surgery (11.4%) compared with those who did not receive clopidogrel (10.5%) [14]. These results differ from ours where patients receiving clopidogrel and ASA within 5 and 2 days, respectively, before surgery had 3-fold increase in risk of re-exploration for bleeding and higher number of blood transfusions with respect to other groups where the dual therapy was discontinued earlier on. Moreover, contrary to the findings that preoperative double antiplatelet therapy reduced cardiovascular and ischaemic events in the setting of ACS, we found that clopidogrel and ASA up to the surgery were associated with a 3-fold increase in the risk of postoperative MI. Previous studies have shown similar results. Kapetanakis et al. [17] found a significant higher incidence in the postoperative MI rate (2.2 vs 0.7%, P < 0.01) and a trend in increased risk of postoperative stroke in patients who received clopidogrel within 7 days than in the control group (2.9 vs 1.7%, P = 0.09). Moreover, Ebrahimi et al. [8] in the ACUTY trial revealed that composite ischaemic events were more frequent in patients undergoing CABG within 5 days of receiving clopidogrel than those who had received clopidogrel >5 days before surgery (14.5 vs 8.8%, P = 0.03). Finally, even though the CURE trial demonstrated a trend towards a reduction in MI, stroke and cardiovascular mortality in patients undergoing CABG exposed to clopidogrel at any time compared with those receiving placebo (14.5 vs 16.2%, respectively; relative risk (RR) 0.89, 95% CI 0.71–1.11), a sub-analysis of this study showed that clopidogrel stopped 5 days prior to CABG was associated with a 3.1% absolute increase in the risk of cardiovascular mortality and composite ischaemic events compared with patients who received clopidogrel >5 days before surgery (11.2 vs 8.1%, RR 1.38, 95% CI 0.92–2.07) [7].

Several other potential predisposing factors such as left main disease, diffuse three-vessel disease, preoperative unstable angina, poor left ventricular EF or reoperations for CABG have been reported to increase the risk of postoperative MI [18]. However, in our study, double antiplatelet therapy up to surgery was the only independent predictor of postoperative MI. The increased risk in postoperative MI in patients under clopidogrel and ASA treatment might be related to the use of blood products transfused for the treatment of bleeding and re-exploration. It has been shown that blood cell transfusion and reoperation for bleeding in patients having cardiac surgery are strongly associated with ischaemic postoperative morbidity and increased mortality [19, 20]. In our study, double antiplatelet treatment up to surgery was associated with a 2-fold increased risk of postoperative blood transfusions. Other predictors were advanced age, gender, pulmonary disease, poor EF, heparin IV, low preoperative haemoglobin and on-pump CABG procedures, also reported by Ferraris et al. [21]. A recent meta-analysis of observational studies showed that recent exposure to clopidogrel before CABG was associated with a higher operative mortality and this was related to blood loss, blood transfusion and re-exploration for bleeding [15]. These results might be due to the functional and structural changes that occur in stored blood cells. The increased red cell aggregability and rigidity, the accumulation of potentially toxic microparticles or proinflammatory cytokines [22] as well as the depletion of NO in the stored blood might promote vasoconstriction, platelet aggregation and ineffective oxygen depletion, causing tissue hypoxia and organ dysfunction [23].

The use of aprotinin may increase the risk of postoperative MI as well. Mangano et al. [24] found that aprotinin was associated with a 55% increase in the risk of MI or heart failure after cardiac surgery. However, Brown et al. [25], analysing 138 randomized trials, showed that aprotinin did not increase the risk of mortality, stroke and myocardial infarction. Our results confirm that aprotinin, although no longer in use, was a predictor of lower blood loss, but not of postoperative MI.

Finally, as shown by Fig. 2, the risk of postoperative MI and re-exploration for bleeding increased with decreasing time from the last dose of clopidogrel and ASA treatment. Specifically, these risks increased exponentially if the patients were still under the effect of clopidogrel and ASA within 5 and 2 days, respectively. Therefore, according to our findings and in line with the European guideline on antiplatelet and anticoagulation management for patients requiring urgent CABG surgery, we recommend to discontinue clopidogrel and ASA at least 5 and 2 days, respectively, before surgery to reduce adverse outcomes such as postoperative MI, re-exploration for bleeding and blood transfusions.

This study has several limitations. It is based on the retrospective analysis of our large institutional observational prospectively collected database and although we performed multivariable analyses to exclude potential confounding variables, it is subjected to bias by unmeasured confounders and the relatively small sample size. Furthermore, in our database no information was recorded about the dose of clopidogrel, clopidogrel resistance, the type of myocardial infarction (STEMI and NSTEMI), the diseased status of the bypassed coronary arteries, the implantation of stents as well as the use of short live glycoprotein (GPIIb/IIa inhibitor) before surgery.

In conclusion, clopidogrel in combination with ASA up to the time of surgery is associated with an increased risk of postoperative MI, reoperation for bleeding and blood loss in patients undergoing CABG. Withholding clopidogrel or ASA for 5 and 2 days, respectively, before surgery and starting bridging therapy with an intravenous short-acting reversible glycoprotein IIb/IIIa inhibitor may avoid these undesirable morbidities, while at the same time prevent the recurrent of cardiovascular adverse events while awaiting surgery. A multicenter prospective randomized trial is required to confirm our data.

Funding

This study was supported by The British Heart Foundation, the Garfield Weston Trust and the NIHR Bristol Biomedical Research Unit in Cardiovascular Medicine.

Conflict of interest: none declared.