3. RESULTS

3.1 Characteristics of the studies
A total of 200 articles were identified by the systematic search (Figure 1), of which 3 randomised controlled trials and 15 observational cohort studies were included in the systematic review. The details of the included studies are summarised in Table 1. All 15 observational studies had a score of more than 6 with NOS, and therefore, are determined as high-quality observational studies. Three randomised controlled trials met at least 6 of the 12 criteria set in the 2015 Updated Method Guidelines for Systematic Reviews in the Cochrane Back and Neck Group10, and they are recognised as high-quality studies.
3.2 Assessment of antiplatelet resistance
The assessment of antiplatelet resistance widely varies using several different platelet function tests and by using downstream arachidonic acid breakdown products such as serum TxB2 or its metabolite in the urine, 11-dehydroTxB2, reflecting the effect of aspirin on platelets.

3.2.1 Light transmission aggregometry:

Platelet-rich plasma (PRP) is prepared by centrifuging the 5ml of anticoagulated blood sample at 150g for 10 minutes at room temperature. PRP is then adjusted to a platelet count of a minimum of 150 000 - 300 000 µl. The samples are then assayed using light transmission aggregometry by adding 0.05 ml of arachidonic acid. However, it can also be performed without adjustment of platelet count, and type I collagen and ADP could also be used instead of arachidonic acid.11 The aggregation was plotted against time and reported as total aggregation % at 5 minutes.12,13

3.2.2 Impedance platelet aggregometry :

It measures platelet aggregation by continuously monitoring electrical impedance changes due to activation of platelets and adhesion to the metal sensor electrodes in 3-5 separate channels in multiple electrode aggregometry (MEA).14–21 Whole blood sample is used in each channel with the addition of arachidonic acid to assess the effect of aspirin (ASPItest ), ADP for platelet P2Y12 inhibitor effect (ADPtest ) or thrombin receptor agonist peptide to measure glycoprotein IIb/IIIa inhibitor effect (TRAPtest ).14,19–21 Collagen can also be used as a substitute for arachidonic acid to assess the effect of aspirin.17 The aggregation result is obtained as an arbitrary area under the curve (AUC) or presented as an aggregation unit against time (AU x min).

3.2.3 Platelet function assay (PFA):

PFA-100 (Dade Behring, Germany) is a commercially available point-of-care platelet function assay, which assesses platelet activation under high shear stress through aspiration of whole blood through collagen/epinephrine coated apertured cartridges (CEPI) or collagen/ADP coated apertured cartridges (CADP).11,22The assessment is reported as aperture closure time (CT), the time taken for subsequent activation of platelets to obstruct the apertures in the CEPI and CADP cartridges.

3.2.4 VerifyNow assay :

VerifyNow system (Accumetrics, San Diego, CA, USA) is a cartridge-based rapid assay system, which assesses the effect of aspirin on platelet reactivity in VerifyNow Aspirin Test using arachidonic acid as an agonist and VerifyNow P2Y12 Test measures the direct inhibition of clopidogrel on the P2Y12 receptors.23,24 Aspirin test results are expressed as aspirin reaction units (ARU) and P2Y12 test results as P2Y12 reaction units (PRUs).

3.2.5 Thromboelastogram (TEG):

Heparinised whole blood is used in the TEG assay (Haemoscope Corp, Niles, IL, USA) to evaluate the platelet function in terms of clot maximum amplitude with added arachidonic acid (MAAA) or without a platelet agonist (MA0), which is compared with kaolin-activated TEG assay (MAKH) to derive percent of platelet aggregation using the formula: %MAAA = [(MAAA - MA0) / (MAKH- MA0)] x 100%.17 The result is reported as a percent aggregation of platelets.

3.2.6 Whole-blood flow cytometry:

The antiplatelet resistance can be measured by incubating blood with or without arachidonic acid ​​(1.0 mmol/L) for 2 minutes, adding radiolabeled antibodies to CD41a or CD62P receptors on platelets, fixing the samples with 1% paraformaldehyde and analysing with a fluorescent cell sorter (Becton-Dickinson FACScan; BD Immunocytometry Systems, San Jose, CA, USA).17 The result is described as the percent increase in the expression of the CD62P receptor after activation.

3.2.7 Thromboxane B2 (TxB2):

The urinary 11-dehydroTxB2 is the excretory form of TxB2 and its concentration is usually measured using enzyme immunoassay kits (Cayman Chemical, MI, USA).11–13 The results are normalised for urinary creatinine concentration. Serum TxB2 level reflects cyclooxygenase-2-dependent thromboxane biosynthesis, and the level is measured in the plasma after whole blood is cultured at 37 °C for 24 hours and centrifuged at 700 x g for 15 minutes.11Serum TxB2 level can also be measured by using immunoassay18 (Neogen, Lexington, KY, USA) or radioimmunoassay.25 Alternatively, centrifuged plasma could be used to measure serum 11-dehydroTxB2 level with an enzyme immunoassay kit (Assay Designs Inc, Ann Arbor, MI, USA).17
3.3 Definition of antiplatelet resistance
There is no uniform definition of antiplatelet resistance in literature, and it varies hugely with each method of assessment. Light transmission aggregometry-derived platelet aggregation of ≥20% with arachidonic acid is regarded as aspirin resistant,13 and in some studies, it is an aggregation of >30%.26,27 Aspirin resistance is determined by impedance aggregometry if AUC ≥30 units15,16,21,28 or AUCASPI>300 units (APSItest value >75 percentile).17,18,20
Using the VerifyNow system, ARU >550 and PRU >230 are considered aspirin resistance and clopidogrel resistance respectively,23,29 although, PRU cut-off point could be as low as 188 for clopidogrel to be resistant.24 Aspirin resistance is determined as having a collagen and/or epinephrine (CEPI) closure time <193 seconds in the PFA-100.22
Serum TxB2 inhibition <90%18 and increase in serum 11-dehydro TxB2 >25% from baseline,17 urinary 11-dehydro TxB2 levels higher than 67.9ng/mmol of creatinine is illustrated as aspirin resistance.12 Platelet aggregation >50% on TEG30,31 and 25% increase in expression of CD62P receptor following simulation in whole-blood flow cytometry are other definitions of aspirin resistance.17
3.4 Antiplatelets used in the studies
All studies used aspirin as primary antiplatelet therapy, and clopidogrel is added to aspirin to constitute dual antiplatelet therapy in some studies.13,18,24,28 While clopidogrel 75 mg is always used, the dosage for aspirin varies from 80-325 mg with 100 mg being the most commonly used dosage. A postoperative loading dose of intravenous aspirin 500 mg was used in one study.27
3.5 Incidence of antiplatelet resistance
The incidence of overall aspirin resistance ranged from 11-51.5%13,15,17,20–23,26–28 and the incidence of clopidogrel resistance was reported as 22%.23 12.6% of patients on dual antiplatelet therapy were found to be resistant to both aspirin and clopidogrel, however, it was reduced to 10.6% after 30-day treatment.23 Preoperative aspirin resistance was found to be 13-29%20,26,27.
In terms of TxB2 measurements, inhibition >90% was not achieved until postoperative day 5 and only 34% of patients reached the effective platelet inhibition by then.12,18Insufficient inhibition of TxB2 is observed with aspirin 100mg, but not with a higher dose of 325mg.11
The aspirin resistance had disappeared in all previously perioperative resistant patients when retested at 6-month13 and 12-month follow-ups.26,27
3.6 Effect of Cardiopulmonary Bypass (CPB)
The effect of CPB on aspirin resistance is not clear. Platelet aggregation and thromboxane are significantly inhibited after off-pump CABG, but not after the on-pump CABG.25 The cardiopulmonary bypass time is described as an independent predictor of an ASA non-response in one study.21 Nonetheless, it was also demonstrated that CPB has no significant effect on aspirin resistance in other studies.26,27
3.7 Outcomes

3.7.1 Vein graft occlusion

Antiplatelet resistance is a predictor of graft occlusion.23 Aspirin resistance, together with compromised endothelial integrity in vein grafts, leads to graft thrombosis and failure within a few days after CABG.17Moreover, the risk of late occlusion of vein grafts is increased by 13 folds (odds ratio) in patients with aspirin resistance.22 Dual antiplatelet therapy with aspirin and clopidogrel is a strong predictor of vein graft patency and is associated with the reduced vein graft occlusion rate.23

3.7.2 Mortality, Myocardial Infarction (MI) and Stroke

There was no overall difference in mortality, MI or stroke in patients with aspirin resistance and those without it at 6-month and 12-month follow-up.20,21,28 Whereas, all patients who died during the follow-up period exhibited aspirin resistance previously.26,27
The addition of clopidogrel to aspirin does not reduce adverse outcomes or increase bleeding episodes. However, dual antiplatelet therapy leads to a lower rate of adverse events in younger (age <65 years) obese patients with body mass index (BMI) >30.28 In patients with clopidogrel resistance undergoing off-pump CABG, high residual platelet reactivity is associated with higher mortality, MI and target vessel revascularisations.24

3.7.3 Postoperative immediate blood loss

Postoperative 12-hour blood loss was higher in preoperative aspirin-sensitive patients compared to the patients with preoperative aspirin resistance (mean volume of 555 ml vs 406ml).29Although the chest drain output was comparable within the first hour after surgery, the aspirin-sensitive group had more blood loss at 6 and 12 hours. In addition, they are more likely to require allogenic blood transfusion postoperatively.14