Role of Aspirin Resistance, Mechanical Factors and Inammation on Early Saphenous Vein Graft Thrombosis After CABG

Objectives: This study was carried out to determine the role of pre-operative and transient aspirin resistance in the formation of early saphenous vein graft (SVG) thrombosis six weeks after coronary artery bypass graft (CABG) surgery and to analyze the other factors, such as mechanical and inammation factors, that are also suspected of contributing to the formation of early thrombosis. Methods: Pre- and post-operative blood samples were taken from 99 subjects, whom 74 patients were undergoing elective on-pump CABG and receiving aspirin as monotherapy, for evaluation of inammation parameters and the state of aspirin resistance using a Platelet Function Analyzer-200 (PFA-200). Transit time ow measurements (TTFM) were performed intra-operatively to determine mechanical factors. Multi-sliced computed tomography (MSCT) was done six weeks after surgery to determine the patency of the vein grafts. Result: In the 222 vein conduits, aspirin resistance was related to early vein graft failure due to thrombosis (p < 0.001; relative risk (RR) = 3.69). The massive increase of interleukin 6 (IL-6) levels after surgery were related to the existence of post-operative transient aspirin resistance (p < 0.001). Transient aspirin resistance (IL-6 > 122.5) was associated with early graft failure (p = 0.029; RR = 8.6) compared to the aspirin-sensitive group (IL-6 > 122.5). Conclusion: Aspirin resistance plays a primary role in early vein graft thrombosis. Transient aspirin resistance accompanied by an increase of inammation factor (IL-6) signicantly increases the risk of early vein graft thrombosis after CABG.


Background
Coronary artery bypass grafting (CABG) surgery is a revascularization modality used in patients with coronary artery disease (CAD). In this procedure, saphenous vein grafts (SVGs) are the most frequently used conduits because they are easy to harvest, have a large diameter, convenient anatomy, and technical features, and can be used for multiple grafts. However, SVG conduits are more susceptible to thrombosis during the rst postoperative year (15-18%), which increases the risk of repeat revascularization, myocardial infarction, and death. 1,2 Aspirin is believed to reduce the incidence of post-CABG thrombosis by suppressing platelet function through cyclooxygenase (COX)-1 inhibition. 3 COX-1 is an enzyme that acts as a catalyst in regulating thromboxane A2 (TXA2) from arachidonic acid (AA). Nonetheless, routine aspirin administration has not been shown to reduce the incidence of early SVG thrombosis after CABG. It is suspected that this phenomenon is closely related to aspirin resistance. 4,5 Aspirin resistance is de ned as inappropriate platelet inhibition that correlates to the persistence of thromboxane formation and high ex-vivo platelet reactivity. In clinical settings, it has been shown that platelet activation and aggregation are not being inhibited, so the formation of thrombosis is ongoing. Page 3/15 There are two types of aspirin resistance: the resistance found before the CABG procedure and the temporary resistance that occurs after CABG, which usually lasts up to 30 days in subjects who were previously sensitive to aspirin. These events could occur as a result of the in ammatory response during CABG (e.g., from the use of a cardiopulmonary bypass (CPB) machine), which includes the increasing platelet count, endothelial cells, leukocyte, complement, and coagulation cascade activation. 1,4,6 Besides aspirin resistance, early SVG thrombosis post-CABG could also be caused by mechanical factors, such as native vessel diameter and conduit blood ow, or by in ammation factors that occur through the formation of transient aspirin resistance post-CABG or directly by in uencing thrombus formation. This SVG thrombosis could trigger future major and minor cardiovascular events. Therefore, this study evaluated the relationship between pre-operative aspirin resistance, transient aspirin resistance, in ammation, and mechanical factors related to the incidence of early SVG thrombosis after CABG.

Subjects And Methods
The study design is an observational cohort study with a repeated measurement conducted in Harapan Kita National Cardiovascular Centre from January 2014 until March 2015. This study has been ethically reviewed and approved by the ethics committee of the National Cardiovascular Centre Harapan Kita. The populations of this study are CAD patients undergoing elective on-pump CABG procedures. Nonprobability consecutive sampling method was used in subjects that meet the inclusion and exclusion criteria.
Inclusion and exclusion criteria The inclusion criteria included subjects with CAD undergoing elective conventional CABG surgery, a creatinine clearance time (CCT) of more than 50 ml/minute, signed informed consent to participate in the study, and a willingness to be monitored for six weeks after surgery. The exclusion criteria were subjects with CAD complicated by heart valve surgery, indications of another procedure beside CABG, intraoperative coronary endarterectomy, a mean graft ow (MGF) of less than 10 ml/sec and/or a pulsatility index (PI) of more than 5, allergies to contrast media, a target vessel diameter of less than 1.0 mm, a CPB duration of more than 120 min, an aortic cross-clamp duration longer than 90 min, the presence of cardiovascular complications after surgery, and the use of extracorporeal membrane oxygenation (ECMO) and/or continuous veno-venous hemo ltration (CVVH) after surgery. This study was conducted on subjects that ful lled all inclusion and exclusion criteria, had been informed about the objectives and the risk of the study and had signed the informed consent.
Aspirin administration and pre-operative assessment All subjects were given 100 mg aspirin once daily as the only anticoagulant for a minimum of seven days before a laboratory test of aspirin sensitivity was done using a PFA-200. The PFA cut-off point was 193 seconds. A value of < 193 seconds was considered as resistant to aspirin while a value of ≥ 193 seconds was considered as sensitive to aspirin. Other pre-operative laboratory tests, such as IL-6, CRP, leukocyte, and other routine hospital tests, were performed simultaneously to obtain baseline data. The aspirin administration was discontinued seven days before surgery.

Intra-operative assessment
The subjects underwent conventional CABG surgery following standard procedures. TTFM, which include MGF and PI, were measured in each SVG intra-operatively to evaluate the quality of the ow and the patency of the conduit. Post-operative assessment At six hours, post-operative, IL-6, CRP, and leukocyte levels were measured, and aspirin was continued when there was no evidence of signi cant bleeding. At 48 hours post-operative, additional leukocyte and CRP measurements were performed. Three days after surgery, the PFA test for aspirin sensitivity was repeated to reassess aspirin resistance status. The subjects were further classi ed into three groups: aspirin resistant, transient aspirin resistant, and aspirin-sensitive. Upon hospital discharge, the subjects were given 100 mg aspirin once daily as the only anticoagulant for a period of weeks. Pill counts were performed in each post-operative appointment. After six weeks, MSCT was performed to assess SVG patency.
Statistical analysis Statistical analyses were performed using IBM SPSS V22.0. Logistic regression was used to assess the relations among the high levels of platelet reactivity, in ammatory response, and early SVG thrombosis.
Results were considered signi cant with a p-value < 0.05. Univariate analyses were performed on a pergraft basis for the odds of occlusion versus patency. A categorical variable was analyzed using chisquare or Fischer's exact tests and was presented in frequency and percentage. Meanwhile, continuous variables were calculated using a T-test or non-parametric test (Mann-Whitney U test).

Result Subject characteristics
A total of 99 subjects and 222 SVGs were included in this study. There were 79 male subjects and 20 female subjects. There were no differences with regards to demographic characteristics or pre-operative risk factors between the patients who had occluded grafts and patents. Intraoperative data, such as ejection fraction, CBP, and cross-clamp time, were also similar ( Table 1). Based on aspirin sensitivity, the subjects were then further classi ed into three different subgroups: resistant (83 conduits), transient (65 conduits), and sensitive (74 conduits).  In this study, IL-6, CRP, and leucocyte were evaluated as representative in ammation factors that contribute to early SVG thrombosis formation.

IL-6
The median pre-operative values of IL-6 were insigni cantly different in each group (patent or occluded; p = 0.204) and subgroup (Table 3). Post-operatively, the mean values in each group and subgroup also increased insigni cantly (p = 0.582) ( Table 3).  (Table   3). However, these median values also increased insigni cantly in each group and subgroup. These results were further analyzed by plotting a receiver operating characteristics 6t (ROC) curve. The graph analysis showed that the validity of the analysis was 55.3% (45.1-65.5%; p = 0.267), and the crosssection coordinate was 101.5, with 48.9% sensitivity and 80% speci city. The analysis showed that a CRP value of more than 101.5 plays a signi cant role (p = 0.004) in the formation of early SVG thrombosis, with a relative risk of 2.11 (Table 4).  The median pre-operative and post-operative leucocyte values were insigni cantly different (p = 0.227 vs. p = 0.547) in each group (Table 3), although there was an increase in postoperative values. The highest increase was achieved 48 hours post-CABG.
Role of IL-6 in the transformation of aspirin sensitivity from sensitive to transient Of the 222 vein conduits, 139 were categorized as aspirin-sensitive before the subject underwent CABG. However, 65 of these subjects transformed into a resistant state three days after the surgery, which was later called a transient group.  were not related to the occurrence of early vein graft occlusions. Furthermore, the transient subgroup alone was also not related to the occurrence of early graft patency (p = 0.191). However, this study sought to analyze both variables, and it was found that SVG occlusion was more prevalent in subjects in the transient group where IL-6 > 122.5 (RR = 8.6, p = 0.029) than in the group where IL-6 ≤ 122.5 (RR = 1.12; p = 0.589) when each was compared to the sensitive group with the same IL-6 values ( Table 6).

Role of mechanical factors in the formation of early vein graft occlusion
The lower intra-operative blood ow group had a greater prevalence of early graft occlusion than those with the higher ow. However, the differences did not reach statistical signi cance (24.4% vs. 17.5%; p = 0.210). The pulsatility index and native vessel diameter also failed to discriminate between the occluded and patent grafts (pulsatility index < 3 with 18.2% occlusion vs. PI 3-5 with 25.9%, p = 0.176; diameter 1-1.5mm with 20.5% vs. D >1.5mm with 21.6%, p = 0.846) ( Table 7). Cox regression was used in performing the multivariate statistical analysis, following the backward stepwise method. There were 10 included variables with p ≤ 0.25, which were suspected of contributing to the occurrence of early vein graft occlusion.
Based on the Wald test results, when the p-value of a variable was > 0.10, the variable had to be eliminated in the next step. However, if the p-value were < 0.10, the corresponding variable would be included in the next step.
As a result, there were two critical variables: PFA category resistant, with p<0.001, and PFA category transient, with p = 0.191, regarding the occurrence of early vein graft failure six weeks after CABG. In this study, the formula used to calculate the probability of early vein graft occlusion six weeks after CABG was as follows: y = -2.259 + 1.637 × PFA Category Resistant + 0.668 × PFA Category Transient The probabilities of early vein graft occlusion six weeks after CABG were 34.9% in the resistant group, 16.9% in the transient group, and 9.5% in the sensitive group. The relative risk of the occurrence of early vein graft thrombosis was calculated by comparing the probabilities of the resistant and sensitive groups, and the result was 3.69 times more likely to develop early graft failure.

Discussion
CABG is a common revascularization modality in patients with advanced CAD that is performed following a standardized general technique. 7 Aspirin is a routine oral antiplatelet given in patients who have undergone CABG in order to prevent/reduce the risk of developing other cardiovascular events. 8 Despite the advances in surgical technique and postoperative management, including the routine administration of acetylsalicylic acid (ASA), the rate of early SVG occlusion remains high, especially in the rst year after surgery (15%). 9,10 Although graft failure in the rst post-operative week is frequently blamed on surgical technique, we minimized this issue as a confounding variable by the routine use of intraoperative ow measurements (TTFM). 3,11−13 Moreover, there is an aspirin non-responsive trait in this study population, which has raised a fundamental question regarding the existence of the relationship between aspirin resistance and the prevalence of early SVG occlusion in patients who have undergone CABG. [14][15][16] The ndings of this study have answered this question and show that there is a strong association between aspirin resistance and early SVG failure. Speci cally, we found that aspirin resistance was signi cantly related to the occurrence of early graft failure, with a relative risk of 3.69 (CI 95% 1.72-7.93). In addition, the transient group was 1.79 times more likely to develop early graft failure compared to the sensitive group, although the result was not statistically signi cant (p = 0.191).
PFA-200 was used in this study because it is simple and reliable; however, there are no laboratory parameters or measurements that have been identi ed as a gold standard to assess aspirin resistance. It is also not included as a standard pre-operative screening because there is not enough evidence to describe its role. Therefore, its speci city and ultimate usefulness are still debatable. 3,13 Furthermore, this study did not assess the potential genetic contribution to the aspirin resistance trait and the formation of early SVG thrombosis.
Based on a previous study by Poston et al., mechanical factors were considered as the most important factors in early SVG failure. 12 Inconsistent with this nding, our study found no difference between MGF and PI in the early SVG occlusion. The lower intra-operative blood ow group had a greater prevalence of early graft occlusion than those with higher ow. However, the differences did not reach statistical signi cance (24.4% vs. 17.5%; p = 0.210). The PI factor also failed to discriminate between the occluded and patent grafts (PI < 3 with 18.2% occlusion vs. PI 3 − 5 with 25.9%; p = 0.176).
With regard to the native vessel diameter factor, this study yielded different results than those of Gluckmann et al., as there was no signi cant difference between occluded and patent grafts (diameter 1-1.5mm with 20.5% vs. D > 1.5mm with 21.6%; p = 0.846).
In addition to aspirin resistance and mechanical factors, this study also attempted to correlate in ammation factors with early graft failure. This review was based on the negative effect of a very high in ammation response from the use of a CPB machine during CABG that hypothetically contributed to this phenomenon. [17][18][19] The previous study by Arazi et al. described the in uence of increasing in ammation markers on the transformation of aspirin resistance from sensitive to resistant after CABG. 20 Similarly, in this study, it was found that high post-operative IL-6 levels play a signi cant role in the transformation of aspirin resistance from sensitive to resistant. Moreover, the transient group had a higher prevalence of early graft occlusion. As a result, very high post-operative IL-6 levels > 122.5 in the transient compared to the sensitive group resulted in a risk up to 8.6 times higher (p = 0.029) compared to a risk 1.12 times higher (p = 0.589) in the group with IL-6 levels < 122.5 in terms of developing early graft failure. However, the result was insigni cant when these factors were analyzed independently.
The other in ammation factor that was shown to have a role in early SVG occlusion was a high postoperative CRP value of > 101.5. Based on these results, the role of in ammation in early SVG occlusion remains inconclusive. Independently, each of the in ammation markers had no signi cant impact on early SVG occlusion. However, upon further analysis, a very high post-operative CRP had a signi cantly higher prevalence of early graft occlusion in comparison to a lower value; the increased prevalence was likely related to the post-operative IL-6 values because high post-operative IL-6 values and not CRP were associated with occlusion in the transient group.

Study limitations
A limitation of this study was not including other in ammation parameters, such as TNF-α and other cytokines, in the analysis. Additionally, this study did not include a genetic examination to determine the role of polymorphism in both the existence of aspirin resistance and the formation of early graft occlusion.
Now that aspirin resistance, IL-6 values, and CRP values have been identi ed as important parameters, additional studies will be needed to determine whether other in ammation and genetic factors also play a role. Further, future emphasis on the study of the effect of anti-in ammatory agents on preventing SVG thrombosis would be bene cial.

Conclusion
We have identi ed that aspirin resistance measured with PFA-200 plays a signi cant role in early vein graft occlusion six weeks after CABG. High levels of IL-6 post-CABG had a signi cant impact on the transformation of aspirin resistance from sensitive to resistant (the transient aspirin group). Postoperative CRP values > 101.5 were associated with early SVG failure. Post-operative IL-6 values of more than 122.5 in the transient subjects were associated with a higher risk of developing early graft occlusion six weeks after CABG.