Keywords
Aspirin - atherosclerosis - coronary artery bypass graft - myocardial infarction -
review - systematic review
BACKGROUND
The long-term success of coronary artery bypass graft (CABG) surgery largely relies
on the persistent patency of the graft conduits. Saphenous vein grafts (SVGs) have
the benefits of being abundant and easy to harvest, but their long-term patency compared
to the left internal mammary artery (LIMA) is poor.[1] For vein grafts, generally, 15%–30% are occluded within 1 year after CABG, and about
50% of these occlusions happen in the first 2 weeks.[1] However, after the first year post-CABG, the annual occlusion rate is 2%–5%. Ten
years after the surgery, approximately one-third of the vein grafts that had been
patent at 1 year remained patent and another third become occluded.[2] Other studies have shown that 12% of vein grafts are occluded within 1 year, 25%
within 5 years, and 50% within 12 years after CABG,[3] and even more studies reported an incidence of one or more total SVG occlusions
to be as high as 41% at 1 year after on-pump CABG.[4],[5],[6],[7],[8],[9],[10] This explains why 3% of participants need a repeat operation within 5 years, 10%
within 10 years, and 25% within 20 years.[11] Hybrid revascularization (LIMA to left anterior descending [LAD], and percutaneous
coronary intervention [PCI] to the other occluded coronaries) is thought to be the
solution to the problem of high rates of vein graft failure.[12],[13] However, the utilization rates have been very disappointing, and vein grafts are
still used for the majority of people.[14] Data on the results of hybrid procedures have been inconsistent, unfortunately.[15] This highlights the importance of continuing to search for the optimal strategy
to improve vein graft latency.
Lack of aspirin (acetylsalicylic acid) prescribed at hospital discharge (discharge
aspirin) was the strongest independent correlate of long-term mortality after CABG
in the land mark SYNTAX trial.[16] Platelet inhibition represents a therapeutic mainstay in treating people with CABG,
and they routinely receive aspirin as a standard treatment for preventing occlusion
and preserving bypass graft surgery benefits,[17] and continue it indefinitely.[18] Furthermore, early postoperative aspirin within 6h following CABG has been reported
to be the best approach for the prevention of vein graft occlusion.[19] Platelet inhibition is associated with a reduced risk of death, reduced ischemic
complications, and improved graft patency.[20],[21],[22],[23] This desired effect of aspirin diminishes the later it is administered.[19] Aspirin is the drug of choice for the prevention of SVG closure in the short term
and is recommended for indefinite use following the procedure due to its benefit in
secondary prevention of death and cardiac events in people with coronary artery disease
(CAD).[18] Despite this benefit, its use for longer than 1 year following CABG does not seem
to improve vein graft patency.[24] Aspirin is effective in reducing further events in people with coronary heart disease;
however, evidence is not conclusive as to which dose is optimal.[19]
Antiplatelet therapy with aspirin had a slight beneficial effect on the patency of
peripheral bypass grafts.[25] However, the debate on dosages of aspirin continues. Some studies showed that there
is a lack of additional benefit with high-dose aspirin but this was accompanied with
an increased risk of bleeding.[26],[27] Residual platelet activity was lower in participants who received aspirin 325 mg
compared to participants who received aspirin 100 mg.[28] Moreover, a single dose of aspirin 325 mg on the first postoperative day may have
a greater inhibitory effect on collagen-induced aggregation than a single dose of
aspirin 100 mg.[29]
The uncertainty of the optimum dose of aspirin after CABG is the main reason why this
review is important. We therefore aimed to evaluate the use of different dose regimens
of aspirin to prevent graft occlusion in people who have undergone coronary artery
bypass grafting.
METHODS
This systematic review and meta-analysis was conducted and reported in accordance
with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)
statement.[30]
Criteria for considering studies for this review
We included all randomized controlled trials (RCTs) (irrespective of language or sample
size) comparing different dosages of aspirin for the purpose of maintaining graft
patency in people who have undergone CABG surgery. We excluded all quasi-randomized
studies, such as those allocating using alternate days of the week or surname of the
participant, as they are not truly randomized and are more prone to bias. We did not
include crossover trials. We also excluded trials that did not study aspirin as the
sole therapy or trials not including a comparison group.
A given participant population was only used once; if the same population appeared
in other trials, we included the article that provided the most complete follow-up
data. We also excluded studies where one or more of the participant groups received
another treatment, such as clopidogrel, because it would be difficult to adjust for
the effects of the additional intervention.
We included trials that enrolled people who had undergone coronary artery surgery
(both on- and off-pump, and including emergency and elective procedures) and had been
placed on aspirin therapy after surgery. We included all trials of participants aged
18 years and older, of either gender, and in any clinical setting.
We included all studies comparing different dosages of aspirin in participants who
have undergone coronary bypass surgery.
We aimed to analyze the outcomes for different lengths of follow-up: up to 3 months,
6 months, 1 year, 5 years, and 10 years, if possible.
Primary outcomes
-
Short-term postoperative cardiovascular-related mortality (i.e., within 30 days of
the operation)
-
Short-term postoperative all-cause mortality (i.e., within 30 days of the operation)
-
Aspirin adverse effects:
-
Minor adverse effects (e.g., gastritis)
-
Major adverse effects (such as gastrointestinal bleeding, acute renal failure)
Secondary outcomes
-
Failed on first CABG attempt
-
Need for coronary intervention
-
Recurrence of cardiovascular events (e.g., myocardial infarction [MI], stable, or
unstable angina)
-
Serum levels of thromboxane B2 (TXB2) as a measure of antithrombotic effects
-
Heart failure
-
Health-related quality of life (as defined by the individual trials)
-
Health-related costs
Search methods for identification of studies
Electronic searches
We searched the following electronic databases: the Cochrane Central Register of Controlled
Trials (CENTRAL), MEDLINE (Ovid), Embase (Ovid), and Web of Science (Thomson Reuters)
from their inception till February 2018.
We used medical subject headings (MeSH) or equivalent and text word terms and imposed
no language restrictions. We also searched the Clinical Trials database (www.clinicaltrials.gov)
and the World Health Organization (WHO) International Clinical Trials Registry platform
(ICTRP) (apps.who.int/trialsearch/) for ongoing trials. Appendix 3 includes the search
strategy for these two databases.
Searching other resources
We checked reference lists of reviews and retrieved articles for additional studies.
We contacted experts in the field for unpublished and ongoing trials, and we contacted
trial authors, where necessary, for any additional information.
Data collection and analysis
We performed the review and meta-analyses following the recommendations of Cochrane.[31] We performed the analyses using Review Manager 5.[32]
Selection of studies
Two authors independently inspected each citation from the searches and identified
relevant abstracts (FA, RZY, AC, AA, and MA). A third author inspected a random 20%
sample of these citations to ensure reliability. Two authors obtained and inspected
each full report of the abstracts that met the review criteria (FA, RZY, AC, AA, MA,
and WA), in addition to citations that the authors disagreed on. A third author inspected
a random 20% of these full reports to ensure reliable selection. Where it was not
possible to resolve a disagreement by discussion, we attempted to contact the authors
of the study for clarification.[31]
Data extraction and management
Two authors independently extracted data from each of the studies (FA and WA). We
discussed and documented any disagreements. With remaining disagreements, a third
author helped clarify issues, and we documented the final decisions. We extracted
data presented only in graphs and figures, whenever possible, but only included them
if two authors independently had the same result. If studies were multicenter, where
possible, we extracted data relevant to each component center separately.
We used a standardized template of a data collection form to extract data on methods,
participants, interventions, and outcomes.
Assessment of risk of bias in included studies
Working independently, two authors (FA and MA) assessed methodological risk of bias
of included studies for adequacy of sequence generation, allocation concealment, blinding
(participants, personnel, and outcome), drop-out rates (incomplete outcome data),
analysis of intention to treat (ITT), selective outcome reporting, and other biases
using the tool described in the Cochrane Handbook for Systematic Reviews of Interventions.[31]
We assessed and categorized the risk of bias in each domain and overall bias as the
following:
-
Low risk of bias: plausible bias unlikely to seriously alter the results
-
High risk of bias: plausible bias that seriously weakens confidence in the results
-
Unclear risk of bias: plausible bias that raises some doubt about the results
When any disagreement arose, we made the final decision by consensus, with the involvement
of another author. We contacted authors of the studies when details about randomization
or other characteristics of the trial were missing. We reported nonconcurrence in
quality assessment, but when disputes arose as to which category a trial was to be
allocated, we obtained resolution by discussion.
Measures of treatment effect
For binary outcomes (e.g., MI or no MI at follow-up), we had planned to calculate
a standard estimation of the random-effects risk ratio (RR) and its 95% confidence
interval (CI). However, none of the dichotomous outcomes were reported by the included
studies, therefore, we had no such data.
For continuous outcomes (e.g., TXB2), we calculated the mean values and standard deviations
for each intervention and comparison group. Whenever the continuous outcome measurement
was similar enough to allow quantitative pooling, we used the mean difference (MD)
and its 95% CI to summarize the results.
We contacted the authors of all the trials, whenever we found missing data.
Assessment of heterogeneity
We considered all included studies without any comparison data to judge clinical and
methodological heterogeneity. We inspected all studies for clearly outlying situations
or people that we had not predicted and discussed them fully.
We visually inspected forest plots to identify trials with nonoverlapping CIs to suggest
the possibility of statistical heterogeneity.
We investigated heterogeneity between studies by considering the I2 statistic and the χ2
P value. The I2 statistic provides an estimate of the percentage of inconsistency thought to be due
to chance.[33] The importance of the observed value of the I2 statistic depends on the magnitude and direction of effects, and the strength of
evidence for heterogeneity (e.g., P value from χ2 test, or a CI for the I2 statistic).
We interpreted an I2 statistic estimate of 50% or greater accompanied by a statistically significant χ2 statistic as evidence of substantial levels of heterogeneity,[33] and explored reasons for heterogeneity. When inconsistency was substantial and we
found clear reasons for this, we planned to present data separately.
Assessment of reporting biases
Owing to the small number of the included studies, we were not able to conduct the
planned assessment of publication and reporting biases.
Data synthesis
We understand that there is no closed argument for preference for use of fixed-effect
or random-effects models. The random-effects method incorporates an assumption that
the different studies are estimating different, yet related, intervention effects.
This often seems to be true, and the random-effects model takes into account differences
between studies even if there is no statistically significant heterogeneity. However,
there is a disadvantage to the random-effects model as it puts added weight onto small
studies, which often are the most biased ones.[34] Depending on the direction of effect, these studies can either inflate or deflate
the effect size.
We analyzed data using a random-effects model and a fixed-effect model. In case of
discrepancy between the two models, we reported both results. Otherwise, we only reported
results from the random-effects model. We planned to analyze data according to the
ITT principle and to present them as RR and risk difference with 95% CI for dichotomous
variables (did not apply as we had no dichotomous outcome data), and as mean values
and standard deviations for continuous variables. We calculated and pooled the MDs
across the included studies between the two groups of low vs. high aspirin dosage.
RESULTS
Results of the search
The flow diagram for study selection is shown in [Figure 1]. We initially identified 5903 references from the databases search and other sources
to go through the abstract screening phase. After excluding 4031 ineligible studies,
a total of 135 studies moved on to the full-text screening level. Eventually, we included
six studies in the qualitative synthesis, and four in the quantitative synthesis (meta-analysis).
No exclusion of any trial was made on the grounds of not reporting on an outcome of
interest.
Figure 1: Flowchart of the screening process
Included studies
None of the included studies reported any of the patient-important outcomes of interest
we initially set out to evaluate. They did, however, report TXB2 levels, a surrogate
outcome for platelet function, which we analyzed quantitatively.
One of the studies we included[35] had an abstract that was published in 2013,[36] and a first version of the study was published in 2015.[37] We chose to keep the 2013 and 2015 reports of this study in the review for informative
purposes as they had different number of patients reported than the full paper published
later. However, we did not include them in the quantitative analysis as to not count
the same patients move than once.
We included three randomized controlled trials with a total number of patients of
122. Mean age of trial participants across the included three trials was 65.63 years,
and 88.67% were male. [Table 1] described the included studies.
Table 1
Characteristicsof the included studies
|
Author
|
Year
|
Study design
|
Total, N
|
Aspirin dosages used
|
Age in years, mean (range)
|
Male, (%)
|
Country
|
TXB2 measured at
|
|
Paikin
|
2015
|
RCT
|
110
|
81 mg OD, 81 mg 4×/day, 325 mg OD
|
65 (NR)
|
82%
|
Canada
|
4 days post-op
|
|
Paikin
|
2017
|
RCT
|
68
|
81 mg OD, 162 mg BID, 325 mg OD
|
65 (NR)
|
88%
|
Canada
|
4 days post-op
|
|
Ivert
|
2017
|
RCT
|
75 (42 included in analysis)
|
75 mg OD, 75 mg BID, 160 mg OD
|
67.2 (NR)
|
41 of 42 (98%)
|
Sweden
|
1 and 3 months post-op
|
|
Brambilla
|
2010
|
RCT
|
49
|
100 mg OD, 325 mg OD
|
64.1 (NR)
|
40 (81.3%)
|
Italy
|
3 and 5 days post-op
|
|
Abstract of Paikin 2013 later published in 2017 (above):
|
|
Paikin
|
2013
|
RCT—abstract
|
100
|
81 mg OD, 325 mg OD, 81 mg 4×/day
|
65 (NR)
|
84%
|
Canada
|
4 days post-op
|
The three randomized trials we included aimed to evaluate the effects of different
dosages of aspirin by measuring the serum TXB2 levels postoperatively. All studies
also evaluated whether multiple-times-a-day aspirin regimen suppressed TXB2 better
than a once-a-day regimen of aspirin. Specifically, the Paikin 2017 trial studied
the different effects of 81 mg once daily (OD), as compared to the 325 mg OD and 162
mg twice daily (BID). TXB2 was measured on postoperative day 4 in all of the three
groups of patients. The patients in this single-center Canadian trial were undergoing
elective or urgent CABG surgery with or without valve surgery.
The Ivert 2017[38] trial evaluated the platelet-inhibition effects of 75 mg OD as compared to 75 mg
BID and 160 mg OD. TXB2 was measured 1 and 3 months postoperatively. This was also
an open label parallel randomized trial of patients undergoing elective CABG only,
therefore, all of them had stable angina pectoris.
Brambilla et al.,[28] on the contrary, evaluated the effect of two different doses (100 and 325 mg OD)
of aspirin on platelet function and TXB2 levels on postoperative day 5.
Excluded studies
We made the decision to exclude 132 full-text articles due to several reasons: 35
due to an intervention that was not of interest, 36 due to a comparator not of interest,
31 due to a study design that was not of interest, 8 due to use of a second intervention,
12 due to outcomes that were not of interest, and 10 due to a patient population that
was not of interest.
Risk of bias in included studies
The included studies had an overall low risk of bias. The only domain of the Cochrane
tool for risk of bias assessment for RCTs that seemed of concern was relating to duration
of follow-up. Although the studies seemed to have a complete enough follow-up, the
duration of this follow-up was very short. For Ivert 2017, the surrogate outcome we
evaluated (TXB2 level) was measured at 1 and 3 months postoperatively. For Paikin
2017, it was at postoperative day 4. For a study by Brambilla et al., it was postoperative day 5.
Detailed assessment and a visual summary of the risk of bias in the included studies
can be seen in [Table 2] and [Figure 2], respectively.
Table 2
Risk of bias assessment of the included studies
|
Study ID
|
Year
|
Selection of patients
|
Ascertainment of exposure
|
Control for confounding
|
Ascertainment of outcome
|
Follow-up long enough
|
Follow-up complete enough
|
Conflicts of interest
|
|
Paikin
|
2015
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
4 days
|
Low risk of bias
|
Low risk of bias
|
|
Paikin
|
2017
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
4 days
|
Low risk of bias
|
Low risk of bias
|
|
Ivert
|
2017
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
1 and 3 months
|
High risk of bias
|
Low risk of bias
|
|
Brambilla
|
2010
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
Low risk of bias
|
5 days
|
High risk of bias
|
Low risk of bias
|
|
Abstract of Paikin 2013 later published in 2017 (above):
|
|
Paikin
|
2013
|
Unclear (abstract)
|
Unclear (abstract)
|
Unclear (abstract)
|
Unclear (abstract)
|
4 days
|
Unclear (abstract)
|
Unclear
|
Figure 2: Risk of bias of the included studies
Allocation (selection bias)
The randomization procedure was clear, and allocation was concealed in the included
studies.
Blinding (performance bias and detection bias)
Blinding was present for participants, investigators, and staff in the included studies.
Incomplete outcome data (attrition bias)
There was no significant loss to follow-up in the included studies. The only exception
was that of Brambilla et al.,[28] where only 49 participants’ data were analyzed of 56 randomized.
Selective reporting (reporting bias)
No selective reporting was detected.
Other potential sources of bias
No other sources of bias were detected.
Effects of interventions
We pooled the MD of serum TXB2 levels comparing a low-dose aspirin (75–100 mg OD)
and a high-dose aspirin (162–325 mg OD). The analysis [Figure 3] showed a pooled MD of 2.00ng/mL (95% CI: 0.72–3.23, participants = 122, studies
= 3) using the random-effects model. Interestingly, using the fixed-effects model
[Figure 4], the pooled MD was also 2.00ng/dL (95% CI: 0.72–3.23, participants = 122, studies
= 3). As the suppression of TXB2 leads to better antithrombotic function, this shows
that the higher dose aspirin helps achieve better antithrombotic activity in patients
who had undergone CABG. Detailed reporting of the serum TXB2 in the included studies
can be found in [Table 3].
Figure 3: Pooling the thromboxane B2 levels across the included studies using the random-effects
model
Figure 4: Pooling the thromboxane B2 levels across the included studies using the fixed effects
model
Table 3
Thromboxane B2 serum levels
|
Study
|
Year
|
Group 1
|
NI
|
Mean 1
|
SD 1
|
Group 2
|
N2
|
Mean 2
|
SD 2
|
Group 3
|
N3
|
Mean 3
|
SD 3
|
|
NR = not reported
|
|
Paikin
|
2015
|
81 mg OD
|
36
|
13.3
|
17.04
|
325 mg OD
|
36
|
3.4
|
3.5
|
81 mg4×/day
|
38
|
1.1
|
1.41
|
|
Paikin
|
2017
|
81 mg OD
|
22
|
4.2
|
4.4
|
325 mg OD
|
23
|
1.9
|
2.8
|
162mg BID
|
23
|
1.1
|
1.5
|
|
Ivert
|
2017
|
75 mg OD
|
1 1
|
3.4
|
2.6
|
75 mg BID
|
14
|
1.8
|
1.3
|
160mg OD
|
17
|
1.6
|
0.9
|
|
Brambilla
|
2010
|
100 mg OD
|
28
|
13.42
|
36.04
|
325 mg OD
|
21
|
8.33
|
10.99
|
-
|
-
|
-
|
-
|
|
Abstract of Paikin 2013 later published in 2017 (above):
|
|
Paikin
|
2013
|
81 mg OD
|
NR
|
1 1
|
18
|
325 mg OD
|
?
|
3.6
|
6.1
|
81 mg4×/day
|
NR
|
1.1
|
1.4
|
As we had a very small number of included studies, we were not able to carry out any
subgroup and sensitivity analyses. The outcomes of cardiovascular and all-cause mortality,
aspirin adverse effects, failed first CABG attempt, need for coronary re-intervention,
recurrence of cardiovascular events, heart failure, health-related quality of life,
and health-related costs were not reported by any of the included studies. The small
number of relevant studies in the literature also prevents the ability to conduct
publication bias assessment.
DISCUSSION
Summary of main results
In this systematic review and meta-analysis, we evaluated the effects of low versus
high dose of aspirin postoperatively in patients who have undergone CABG. We found
that high-dose aspirin (160–325 mg OD) was superior to low-dose aspirin (75–100 mg
OD) in suppressing serum TXB2 by 2.00ng/mL (95% CI: 0.72–3.23). As this is a surrogate
biochemical outcome, it is unclear if this significant biochemical difference translates
into an important clinical difference in any of the clinical outcomes we had set out
to evaluate. When faced with uncertainty about a surrogate outcome, it is suggested
to investigate the association between the surrogate and clinical outcomes.[39] It has been shown that serum TXB2 was an independent risk factor for vein graft
thrombosis after CABG surgery.[5] Therefore, it is plausible that high-dose aspirin may be more protective.
Overall completeness and applicability of evidence
None of the identified studies in the literature reported any clinical outcomes of
interest. Future studies need to focus on relevant and patient-important outcomes
to better guide medical care for patients undergoing CABG. Also, the outcome of serum
TXB2 was measured at different times by Ivert 2017 (1 and 3 months postoperatively)
as compared to Paikin 2017 (fourth postoperative day) and Brambilla 2010 (fifth postoperative
day). This could be another factor for better consistency in the methodology of outcome
reporting for future studies. The patient samples of the included studies comprised
an overwhelming male majority (89%). This limits the generalizability of the findings
to all patients undergoing CABG. Therefore, future studies should attempt to capture
more balanced and representative samples.
Quality of the evidence
The quality of this evidence, however, is low. This was mainly due to the indirectness
of the outcome measurements (TXB2 is a surrogate outcome) and heterogeneity in the
timing of outcome measurement.
Potential biases in the review process
The main challenge we faced in this systematic review was the lack of evidence to
adequately answer this question. None of the clinical outcomes we had set out to evaluate
were reported in this body of literature at this point. Although reluctant, we had
no other choice but to evaluate a surrogate biochemical outcome that was reported
by the included studies. Surrogate and biochemical outcomes have their own inherent
limitations of insufficiently informing medical practice and patients’ goals of medical
care. It is important to note that although this meta-analysis found a statistically
meaningful MD in TXB2 levels between the low- and high-dose aspirin after CABG, it
is unlikely that this translates to a clinically meaningful difference. Additionally,
although three of the included trials had evaluated the serum TXB2 levels on the fourth
and fifth postoperative day, the third one evaluated it 1 and 3 months postoperatively.
Although the results were not very different, but this casted doubt on the validity
of pooling this study with the other two. One other limitation was the very small
number of studies that were found. We interpreted the findings in light of these limitations.
Agreements and disagreements with other studies or reviews
Significant risk factors for SVG thrombosis within 6 months of CABG surgery in people
taking postoperative aspirin include small target vessel diameter, female gender,
and low mean graft blood flow.[40] Aspirin inhibits platelet activation by irreversibly acetylating platelet COX-1
and preventing the formation of thromboxane A2. Although aspirin has a half-life of
approximately 20min, it produces almost complete inhibition of thromboxane A2 synthesis
for 24h when given once daily.[41] In most patients receiving chronic once-daily aspirin therapy, platelet COX-1 activity
is restored at a rate of approximately 10% per day, reflecting the 10-day platelet
life span and the entry into the circulation of 10% of newly formed platelets. TXB2
is the stable metabolite of thromboxane A2 and arachidonate-induced platelet aggregation.
The 2011 American College of Cardiology (ACC)/American Heart Association (AHA) guidelines
recommend that every person receives daily aspirin therapy after CABG (Class I indication).[18] The 2012 American College of Chest Physicians (ACCP) guidelines on the primary and
secondary prevention of cardiovascular disease also stated that people who undergo
CABG should be started on aspirin and it should be continued indefinitely.[42] Several studies assessed whether early treatment with aspirin inhibits platelet
aggregation, has an effect on graft patency, or improves survival after coronary bypass
surgery. Many of these studies have showed that early use of aspirin after CABG reduces
the risk of death and ischemic complications.[21],[43],[44],[45] However, the guideline and many of the studies took no account of the wide variation
in aspirin doses (from 75 to 325 mg). As a result, low-dose aspirin (75 to 150 mg)
is usually prescribed despite the lack of direct comparison with medium- or high-dose
regimens.[46]
There have been many advances in surgical techniques, interventional cardiovascular,
and medical care for patients with CAD, particularly after CABG surgery. Furthermore,
the use of aspirin, with or without other antiplatelets, has become routine practice
for patients after CABG. However, it is still unclear what optimal aspirin dose should
be prescribed to these patients postoperatively. With long-term success of this surgical
intervention relying on graft patency, finding the aspirin dose most effective to
achieve best clinical outcomes is paramount. Very little evidence exists to fill this
gap, unfortunately. Clinical trials evaluating different dosages of aspirin after
CABG that are evaluating clinical and patient-important outcomes are needed to fill
this gap in knowledge to better inform medical practice.
AUTHORS’ CONCLUSIONS
Implications for practice
High-dose aspirin (160–325 mg OD) is superior to low dose (75–100 mg OD) in suppressing
serum TXB2 levels, therefore, achieving better postoperative antithrombotic function
for patients undergoing CABG.
Implications for research
Clinical trials evaluating different dosages of aspirin after CABG that are evaluating
clinical and patient-important outcomes are needed to fill this gap in knowledge and
to better inform medical practice.
