Sage Journals: Discover world-class research

Abstract

Background:

The administration of aspirin is traditionally discontinued prior to coronary artery bypass grafting (CABG), given a potential risk of excessive postoperative bleeding. Few studies have previously suggested the benefits of continuing aspirin until the time of surgery. The primary aim of this review is to evaluate the effects of preoperative aspirin therapy on several clinically important outcomes in patients undergoing CABG.

Methods:

A meta-analysis of eligible studies of patients undergoing CABG, reporting preoperative aspirin in comparison with no aspirin/placebo and our outcomes, was carried out. The safety outcomes included postoperative bleeding, packed red blood cell (PRBC) transfusion requirements, and reoperation for bleeding. The efficacy outcomes included perioperative myocardial infarction (MI), cerebrovascular accidents (CVAs), and mortality.

Results:

In 8 randomized controlled trials (RCTs; n = 1538), preoperative aspirin increased postoperative bleeding (difference in means = 132.30 mL; 95 % confidence interval [CI] 47.10-217.51; P = .002), PRBC transfusion requirements (difference in means = 0.67 units; 95% CI 0.10-1.24; P = .02), and reoperation for bleeding (odds ratio [OR] = 1.76; 95% CI 1.05-2.93; P = .03). In 19 observational studies (n = 19551), preoperative aspirin increased postoperative bleeding (difference in means = 132.74 mL; 95% CI 45.77-219.72; P = .003) and PRBC transfusion requirements (difference in means = 0.19 units; 95% CI 0.02-0.35; P = .02) but not reoperation for bleeding (OR = 1.13; 95% CI 0.91-1.42; P = .27). Subgroup analyses for RCTs demonstrated that aspirin given at doses ≤ 100 mg/d might not increase the postoperative bleeding, and the dose of 325 mg/d might not be a cutoff value that has clinical and statistical significance. No statistically significant differences in the rate of perioperative MI, CVAs, or mortality were seen between the 2 groups.

Conclusions:

Preoperative aspirin therapy is associated with increased postoperative bleeding, PRBC transfusion requirements, and reoperation for bleeding in patients undergoing CABG. Doses lower than 100 mg/d may minimize the risk of bleeding. Additional RCTs are needed to assess the effects of preoperative aspirin on the safety and efficacy outcomes in patients undergoing CABG.

Keywords

aspirin coronary artery bypass graft safety efficacy

Introduction

Aspirin (acetylsalicylic acid [ASA]) is an integral pharmacological therapy for patients with coronary artery disease. More than 200 clinical trials presented the evidence that aspirin therapy reduced the risk of the combined outcome of serious vascular events by about one-fourth; the risk of nonfatal myocardial infarction (MI) was reduced by one-third, nonfatal stroke by one-fourth, and vascular mortality by one-sixth.¹ The administration of early aspirin after coronary artery bypass grafting (CABG) is indicated to prevent graft occlusion, reduce ischemic events (MI and stroke), and improve survival.² The 2011 American College of Cardiology Foundation (ACC)/American Heart Association (AHA) guidelines recommend that all patients receive daily aspirin therapy after CABG (Class I indication).³ However, there has been controversy regarding preoperative aspirin use due to a potential increased risk of bleeding complications.⁴ Most centers of cardiac surgery are currently in favor of aspirin withdrawal prior to CABG in patients without acute coronary syndrome,⁵ in accordance with current guidelines from the Society of Thoracic Surgeons⁶ and ACC/AHA.⁷ Concerns still exist that abrupt discontinuation of chronic aspirin therapy may trigger a prothrombotic rebound phenomenon, potentially placing patients with CABG at excess risk of ischemic events in the perioperative period.⁸ Evidence from randomized trials and observational studies has suggested the potential for increased postoperative bleeding, transfusion requirements, and reoperation for bleeding attributable to preoperative aspirin therapy.^9,10 However, several reports support the benefits of continuing aspirin until the day of surgery to decrease graft occlusion¹¹ and to improve survival^12,13 and that low-dose aspirin does not increase postoperative bleeding.^14,15

We undertook the present systematic review and meta-analysis of the relevant literature to evaluate the effects of preoperative aspirin therapy on the safety outcomes of the postoperative bleeding, packed red blood cell (PRBC) transfusion requirements, reoperation for bleeding, and the efficacy outcomes of perioperative MI, cerebrovascular accidents (CVAs), and mortality in patients undergoing CABG.

Methods

Eligibility Criteria

We sought eligible studies which met the following criteria: (1) published randomized controlled trials (RCTs) or observational studies; (2) involving patients older than 18 years; (3) undergoing first-time elective CABG, either on- or off-pump, with or without combined procedures; (4) treatment groups discontinuing aspirin less than 7 days before CABG versus no aspirin/placebo; and (5) end points included in the studies (safety end points: postoperative bleeding, reoperation for bleeding, and PRBC transfusion requirements; efficacy endpoints: perioperative MI, CVAs, and mortality). Non-English language studies were excluded as were case reports, editorial comments, and review articles. Only studies or subgroups of patients with stable coronary artery disease were included in the present analysis. Studies or subgroups of patients who were preoperatively exposed to other anticoagulants (such as heparin and warfarin) or antiplatelet drugs (such as clopidogrel and dipyridamole) were excluded in order to eliminate the potential for confounding.

Search Strategy

MEDLINE, EMBASE, and Cochrane libraries were independently searched by 2 reviewers (Xiaochun Ma and Chi Ma) for potentially eligible studies from the earliest available date to May 2013. The search terms used either alone or in combination included CABG, coronary artery bypass graft, coronary artery bypass surgery, aspirin, acetylsalicylic acid, ASA, retrospective, prospective, RCTs, observational studies, patency, occlusion, and bleeding. Reference lists from retrieved articles were also manually searched for articles meeting our criteria.

Abstracts identified using our search strategies were reviewed separately by 2 reviewers (Xiaochun Ma and Yan Yun). The full-text articles that potentially met the criteria were then reviewed in duplicate to determine the inclusion in the analysis. Disagreements between reviewers were resolved by consensus among all the authors.

Data Extraction

Data extraction was performed independently by 2 reviewers (Xiaochun Ma and Chi Ma). Extracted data included study design, assessment of study quality, baseline characteristics, patient demographics, aspirin doses, and safety and efficacy end points. The quality of studies was assessed with the Newcastle-Ottawa scale for observational studies¹⁶ and with the Jadad scale for RCTs.¹⁷ The Newcastle-Ottawa scale assesses the study on 3 aspects, selection of the study cohort, comparability of the cases and controls on the basis of study design, and the methods used to determine exposure to the event studied. The Jadad scale uses a set of 5 criteria to determine the quality of a RCT with a score ≥3 denoting high quality of the study. The meta-analysis has been reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Data Analysis

Review Manager 5.1 and Stata 12.0 were used for performing the meta-analysis. Random-effect models were chosen for pooling the results from included trials.¹⁸ Differences in postoperative bleeding and transfusions were summarized as the mean differences (MDs) of the amounts of postoperative mediastinal drainage of blood (mL) and transfused PRBC (units) between treatment groups, respectively. Reoperation for bleeding, perioperative MI, CVAs, and mortality were summarized in the form of odds ratios (ORs) with 95% confidence intervals (CIs). A 2-sided P value of <.05 was considered statistically significant. Chi-square test of heterogeneity with statistical significance set at α = .10 was used to test for heterogeneity. We also used the I² statistic, which estimates the percentage of total variation across studies, to estimate the inconsistencies between included studies. Publication bias was assessed using the Funnel plots and Egger test with a P value of <.05 considered as the level of significance. Agreement assessment between the reviewers was evaluated using the Cohen k statistic.

Subgroup analysis was carried out to explore heterogeneity between the groups. In the present analysis, the included studies were conducted over a period of more than 30 years and the sample sizes of these studies varied. It was hypothesized that there would be heterogeneity between studies by date of publication and sample sizes. Thus, subgroup analyses for RCTs or observational studies were performed on the basis of sample sizes (large sample size: ≥100 patients; small sample size: <100 patients) and publication year (before 1990; 1990 and after). We also hypothesized a priori that patients given higher doses of aspirin would have increased amounts of postoperative bleeding.¹⁹ Thus, we also explored heterogeneity for RCTs by aspirin doses (cutoff doses: 325 mg/d or 100 mg/d) and by quality based on Jadad scale (high quality: ≥3; low quality: <3).¹⁷ The intervention of aspirin groups was the preoperative intake of aspirin within 7 days of surgery. For individual trials with no events in 1 or both the groups, a continuity correction of .5 was added to each cell for each effect measure.²⁰ When multiple groups were present in a single study, those groups solely receiving aspirin were allowed for inclusion in the present study, and the sample size of the control group was split evenly by the number of included aspirin groups.²¹

Results

Eligible Studies

A total of 5468 potential eligible studies were identified with the literature search. The titles of these studies were reviewed, and 5237 were rejected leaving 231 potentially eligible studies. After review of abstracts, an additional 188 were rejected and 43 studies (12 RCTs and 31 observational studies) were then retrieved in full text to determine the inclusion. Initially, assessment agreement of study selection of the full-text articles between the 2 reviewers led to a k score of .83 (95% CI 0.77-0.89). Then, a further discussion was conducted among all the authors with regard to the studies upon which the 2 reviewers disagreed. Finally, corporate review of full-text articles excluded 3 studies for not reporting our end points, 3 studies for inclusion of other studies within their manuscript, and 10 studies for not providing adequate and detailed statistics (Figure 1).

Figure 1.

Flow diagram depicting the methodology of article selection. RCTs indicate randomized controlled trials.

Study Characteristics

Ultimately, 8 RCTs^{22

–29} involving a total of 1538 patients and 19 observational studies^{12,14,15,30

–45} comprising a total of 19 551 patients met our criteria and were included in this meta-analytical review. Characteristics of the included studies are detailed in Tables 1 and 2 . Based on the Jadad scores, 5 RCTs were of high quality while the other 3 were of low quality. All the 5 high-quality RCTs were published after 1990 with large sample sizes. The 3 low-quality RCTs were published before 1990 with small sample sizes. Of the observational studies, 13 with large sample sizes were published after 1990. The other 6 observational studies had small sample sizes. Of the RCTs, 4 reported aspirin doses ≥325 mg/d, while the remaining 4 RCTs reported doses <325 mg/d; 2 RCTs reported dosages ≤100 mg/d. Aspirin doses in the RCTs differed from 80 to 2600 mg/d, and aspirin doses in the observational studies ranged from 75 to 2400 mg/d.

Table 1.

Outline of RCTs Included in the Meta-Analysis.^a

	High Quality, Large Sample Size and After 1990					Low Quality, Small Sample Size and Before 1990
Characteristics	Deja	Morawski	Kallis	Hockings	Goldman	Ferraris	Karwande	Fuller
Year	2012	2005	1994	1993	1991	1988	1987	1985
Country	Poland	Poland	UK	AUS	USA	USA	USA	USA
Sample size	783	102	100	102	351	34	36	30
Jadad score	4/5	4/5	4/5	4/5	4/5	2/5	1/5	1/5
Blinded	Yes	Yes	Yes	Yes	Yes	No	No	No
Design	Control: 396 ASA: 387	Control: 51 ASA: 51	Control: 50 ASA: 50	Control: 52 ASA: 50	Control: 175 ASA: 176	Control: 18 ASA: 16	Control: 10 ASA: 14	Control: 9 ASA (G2): 11 ASA (G3): 10
Mean age	59	61	62	60	60	62	63	57
Male, %	78	85	81	93	100	88	71	Unknown
ASA dosages	300 mg/d	300 mg/d	300 mg/d	100 mg/d	325 mg/d	325 mg/d	80 mg/d	325-2600 mg/d
Postoperative drugs	ASA: 300 mg/d all patients	ASA: 150 mg/d all patients	Unknown	ASA: 100mg/d ASA group	ASA: 325 mg 6 hours postoperatively all patients	Unknown	Unknown	Unknown
Antifibrinolytics	A portion of patients	No	Unknown	Unknown	Unknown	Aminocaproic acid used in 25% patients of ASA group	Unknown	Unknown
Safety outcomes	Bleeding transfusion reoperation	Bleeding transfusion reoperation	Reoperation	Bleeding transfusion reoperation	Bleeding reoperation	Bleeding transfusion reoperation	Bleeding	Bleeding reoperation
Efficacy outcomes	MI, CVAs, mortality	MI, mortality	MI, mortality			MI, mortality		Mortality

Abbreviations: ASA, aspirin; RCT, randomized controlled trial; MI, myocardial infarction; CVAs, cerebrovascular accidents.

^a Study quality has been assessed using the Newcastle-Ottawa scale (Selection = S; Comparability = C; Exposure = E) and Jadad scale for observational and randomized controlled trials, respectively. Subgroups of patients who were preoperatively exposed to other anticoagulants or antiplatelet drugs were excluded.

Table 2.

Outlines of Observational Studies Included in the Meta-Analysis.^a

	Small Sample Size (Before or After 1990)					Large Sample Size and After 1990
	Michelson	Torosian	Gerrah	Gerrah	Gerrah	Taggart	Rawitscher	Reich	Vuylsteke	Jakics
Year	1978	1978	2003	2004	2005	1990	1991	1994	1997	1999
Country	USA	USA	Israel	Israel	Israel	Scotland	USA	USA	UK	USA
Sample size	25	98	20	94	32	202	100	197	144	177
Design	Control: 16; ASA: 9	Control: 64; ASA: 9	Control: 10 ASA: 10	Control: 48; ASA: 46	Control: 18; ASA: 14	Control: 101; ASA: 101	Control: 72; ASA: 28	Control: 110; ASA: 87	Control: 58; ASA: 86	Control: 49; ASA: 51
NOS	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3
Mean age	54	54	63	69	63	57	61	68	61	64
Male, %	100	89	75	78	72	83	76	68	81	72
ASA dosages	600-2400 mg/d	600-2400 mg/d	100 mg/d	100 mg/d	100 mg/d	75-300 mg/d	85-325 mg/d	Unknown	≤325 mg/d	325 mg/d
Antifibrinolytics	Unknown	Unknown	Unknown	20% of ASA group 29% of control group aprotinin perioperatively	Unknown	Unknown	Unknown	No	No	No
Safety outcomes	Bleeding transfusion	Bleeding reoperation	Bleeding transfusion reoperation	Bleeding transfusion	Bleeding transfusion	Reoperation	Bleeding transfusion reoperation	Bleeding transfusion reoperation	Bleeding transfusion reoperation	Bleeding transfusion
Efficacy outcomes			Mortality	Mortality	MI		Mortality

Table 2.

(Continued) Large Sample Size and After 1990.

	Large Sample Size and After 1990
	Weightman	Ray	Bybee	Srinivasan	Kamran	Gulbins	Hijazi	Jacob	Mikkola
Year	2002	2003	2005	2003	2008	2009	2011	2011	2012
Country	AUS	Canada	USA	UK	Pakistan	Germany	Jordan	USA	Finland
Sample size	797	659	1636	340	30	10900	756	3038	306
Design	Control: 187; ASA (G1): 140; ASA (G2): 255; ASA (G3): 215	Control: 497; ASA: 105	Control: 320; ASA: 1316	Control: 170; ASA: 170	Control: 15; ASA: 15	Control: 8502; ASA: 2398	Control: 356; ASA: 400	Control: 1519; ASA: 1519	Control: 153; ASA: 153
NOS	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S3; C1; E3	S4; C1; E3	S4; C1; E3
Mean age	62	65	69	64	53	66	64	65	68
Male, %	83	76	75	74	90	Unknown	65	78	84
ASA dosages	Unknown	Unknown	Unknown	75-300 mg/d	Unknown	Unknown	100 mg/d	Unknown	100 mg/d
Antifibrinolytics	No	Unknown	Tranexamic acid all patients perioperatively	No	Unknown	Aprotinin all patients perioperatively	Unknown	Unknown	No
Safety outcomes	Transfusion reoperation	Transfusion	Reoperation		Transfusion	Bleeding transfusion reoperation	Transfusion	Reoperation	Bleeding transfusion reoperation
Efficacy outcomes	Mortality		CVAs, mortality	CVAs		MI, mortality	MI	MI, CVAs, mortality	CVAs, Mortality

Abbreviations: ASA, aspirin; MI, myocardial infarction; CVAs, cerebrovascular accidents; NOS, Newcastle-Ottawa scale; S, selection; C, comparability; E, exposure.

Safety Outcomes

Postoperative Bleeding

Randomized controlled trials

Pooling the results of 7 studies showed that preoperative aspirin increased postoperative bleeding (MD = 132.30 mL; 95% CI 47.10-217.51; P = .002). Heterogeneity was not statistically significant (P = .21; I² = 27.5%) and remained not evident when studies were analyzed separately in subgroup analysis by sample size, date of publication, and Jadad score. The 4 high-quality studies, published after 1990 and with large sample sizes, showed a significant increase in postoperative bleeding in the aspirin groups (MD = 105.15 mL; 95% CI 28.40-181.91; P = .01; P heterogeneity = .29; I² = 20.2%). The other 4 low-quality studies published before 1990 and with small sample sizes demonstrated a trend toward increased postoperative blood loss in the aspirin groups (MD = 233.25 mL; 95% CI −15.04-481.53; P = .07; P heterogeneity = .16; I² = 41.5%). But this finding did not reach statistical significance. Heterogeneity between the subgroups was not significant (P = .68; Figure 2). In addition, sensitivity analysis was performed for the RCTs excluding Fuller et al’s that reported an aspirin dose of more than 2000 mg/d. The results were consistent before and after excluding the study of Fuller et al (MD = 122.67 mL; 95% CI 33.28-212.05; P = .01; P heterogeneity = .19; I² = 33.3%).

Figure 2.

Meta-analysis of postoperative bleeding for RCTs (including subgroup analysis according to date of publication, sample size, and Jadad scale). RCTs indicate randomized controlled trials; CI, confidence interval.

Additionally, the studies were analyzed separately according to aspirin dose. For the cutoff value of 325 mg/d, the studies receiving aspirin doses ≥325 mg/d showed a significant increase in postoperative bleeding (MD = 229.61 mL; 95% CI 18.73-440.49; P = .03; P heterogeneity = .22; I² = 32.1%). The studies of patients treated with doses <325 mg/d also demonstrated a significant increase in postoperative blood loss (MD = 101.61 mL; 95% CI 15.80-187.43; P = .03; P heterogeneity = .28; I² = 21.4%). Heterogeneity between the subgroups was significant (P = .02; Figure 3). For the cutoff value of 100 mg/d, the studies receiving doses >100 mg/d showed an increased postoperative bleeding (MD = 160.70 mL; 95% CI 43.13-278.27; P = .01; P heterogeneity = .02; I² = 43.5%). However, the studies on patients treated with doses ≤100 mg/d did not show any significant difference between the 2 groups (MD = 103.08 mL; 95% CI −65.95-272.11; P = .23; P heterogeneity = .38; I² = 0%). Heterogeneity between the subgroups was not significant (P = .92).

Figure 3.

Subgroup analysis of postoperative bleeding for RCTs according to aspirin doses (325 mg/d). RCTs indicate randomized controlled trials; CI, confidence interval.

Observational studies

Pooling the results of 11 studies showed significantly greater postoperative bleeding in the aspirin groups (MD = 132.74 mL; 95% CI 45.77-219.72; P = .003). High heterogeneity was noted (P < .001; I² = 88.3%), thus the subgroup analysis was performed according to sample size and date of publication. In all, 5 studies were published before 1990 and had small sample sizes. The pooled analysis demonstrated an increase in postoperative blood loss in the aspirin groups without heterogeneity (MD = 232.57 mL; 95% CI 147.74-317.41; P < .001; P heterogeneity = .23; I² = 28.4%). The other 6 studies were published after 1990 and had large sample sizes. However, the pooled analysis did not show a significant increase in postoperative bleeding in the aspirin groups (MD = 46.53 mL; 95% CI −39.85-132.92; P = .29; P heterogeneity < .001; I² = 86.1%: Figure 4). Heterogeneity was still statistically significant, and sensitivity analysis was thus performed excluding that of Gulbins et al’s, because the calculated MD for this study was extremely lower (MD = −75.00 mL, 95% CI −109.60-−40.40). The pooled analysis showed that the end point and heterogeneity changed greatly before and after exclusion of Gulbins et al’s study (MD = 75.92 mL; 95% CI 37.60-114.24; P < .001; P heterogeneity = .54; I² = 0%), suggesting the instability of the original results. Heterogeneity between the subgroups was significant (P < .001). In addition, sensitivity analysis was performed for the observational studies excluding the studies of Michelson et al and Torosian et al that reported aspirin doses of more than 2000 mg/d. The results were consistent before and after excluding the studies of Michelson et al and Torosian et al (MD = 97.75 mL; 95% CI 12.62-182.87; P = .02; P heterogeneity < .001; I² = 88.7%).

Figure 4.

Meta-analysis of postoperative bleeding for observational studies (including subgroup analysis according to date of publication and sample size). CI indicates confidence interval.

Packed Red Blood Cell Transfusion Requirements

Pooling the results of 4 RCTs showed significantly greater PRBC transfusion requirements in the aspirin groups (MD = 0.67 units; 95% CI 0.10-1.24; P = .02). Heterogeneity was evident (P heterogeneity = .01; I² = 76.7%) and was not eliminated when the subgroup analysis was carried out by date of publication, sample size, and Jadad score (data not shown). Pooling the results of 14 observational studies demonstrated that preoperative aspirin increased PRBC transfusion requirements in the aspirin groups (MD = 0.19 units; 95% CI 0.02-0.35; P = .02). Heterogeneity was statistically significant (P heterogeneity = .01; I² = 53.4%; Figure 5) and remained evident when studies were analyzed separately according to date of publication and sample size (data not shown).

Figure 5.

Meta-analysis of packed red blood cell transfusion for observational studies. CI indicates confidence interval.

Reoperation for Bleeding

Randomized controlled trials

The results of 7 studies were pooled and showed that preoperative aspirin increased the rates of reoperation for bleeding in the aspirin groups (OR = 1.76; 95% CI 1.05-2.93; P = .03). Heterogeneity was not statistically significant (P = .80; I2 = 0%) and remained not evident when studies were analyzed separately in a subgroup analysis according to sample size, date of publication, and Jadad score. The 5 high-quality studies, published after 1990 and with large sample sizes, showed a significant increase in the rates of reoperation for bleeding in the aspirin groups (OR = 1.74; 95% CI 1.03-2.96; P = .04; P heterogeneity = .61; I2 = 0%). The other 3 low-quality studies, published before 1990 and with small sample sizes, did not demonstrate any difference (OR = 1.99; 95% CI 0.27-14.49; P = .50; P heterogeneity = .57; I2 = 0%), and the heterogeneity between subgroups was not significant (.63; Figure 6).

Figure 6.

Meta-analysis of reoperation for bleeding for RCTs (including subgroup analysis according to date of publication, sample size, and Jadad scale). RCTs indicate randomized controlled trials; CI, confidence interval; OR, odds ratio.

Observational studies

Pooling the results of 11 studies showed no difference in the rates of reoperation for bleeding between the aspirin groups and the control groups (OR = 1.13; 95% CI 0.91-1.42; P = .27; P heterogeneity = .94; I² = 0%; Figure 7).

Figure 7.

Meta-analysis of reoperation for bleeding for observational studies. CI indicates confidence interval; OR, odds ratio.

Efficacy Outcomes

Myocardial Infarction

Pooling the results of either 4 RCTs or 4 observational studies alone showed no significant difference in the rates of perioperative MI between the 2 groups (RCTs: OR = 0.73; 95% CI 0.37-1.44; P = .36; P heterogeneity = .88; I² = 0%; observational studies: OR = 1.24; 95% CI 0.92-1.66; P = .16; P heterogeneity = 1.00; I² = 0%; Figure 8).

Figure 8.

Meta-analysis of myocardial infarction for RCTs and observational studies. RCTs indicate randomized controlled trials; CI, confidence interval; OR, odds ratio.

Cerebrovascular Events

Preoperative aspirin did not significantly decrease the rates of perioperative CVAs, showed by a pooled analysis of 5 studies (OR = 0.80; 95% CI 0.40-1.59; P = .52; P heterogeneity = .19; I² = 34.8%; Figure 9).

Figure 9.

Meta-analysis of cerebrovascular events for RCTs and observational studies. RCTs indicate randomized controlled trials; CI, confidence interval; OR, odds ratio.

Mortality

Pooling the results of either 5 RCTs or 8 observational studies alone did not demonstrate any significant difference in the rates of postoperative mortality (RCTs: OR = 1.04; 95% CI 0.39-2.74; P = .94; P heterogeneity = .99; I² = 0%; observational studies: OR = 0.87; 95% CI 0.66-1.14; P = .32; P heterogeneity = .48; I² = 0%; Figure 10).

Figure 10.

Meta-analysis of mortality for RCTs and observational studies. RCTs indicate randomized controlled trials; CI, confidence interval; OR, odds ratio.

Publication Bias Analysis

The funnel plots and results of Egger tests showed that there was no statistical publication bias observed in all our analyses except for the outcome of MI in the RCTs (Egger test P = .02). It is difficult to interpret this in regard to MI, because there was no significant difference in MI likely due to lack of power.

Discussion

The present systematic review and meta-analysis were performed to assess the effects of preoperative aspirin therapy on the safety and efficacy outcomes in patients undergoing CABG. For safety outcomes, our analysis demonstrated that preoperative aspirin increased postoperative bleeding, transfusion requirements, and reoperation for bleeding, which were demonstrated by pooling the results of RCTs. In 19 observational studies, preoperative aspirin increased postoperative bleeding and PRBC transfusion requirements but not reoperation for bleeding. Subgroup analysis for RCTs by date of publication, sample size, and Jadad score yielded consistent results for postoperative bleeding, suggesting the robustness of the initial analysis. Sensitivity analysis for observational studies was performed excluding the study of Gulbins et al after subgroup analysis by date of publication and sample size in the analysis of postoperative bleeding. The pooled estimate and heterogeneity altered significantly before and after it was excluded, suggesting that the robustness of the subgroup analysis (for all the 6 recent observational studies with large sample sizes) was limited and the results were greatly influenced by Gulbins et al. We failed to completely eliminate the heterogeneity. The difference between Gulbins et al’s and the others accounted for a portion of heterogeneity that could not be removed by simply excluding the study of Gulbins et al. Further investigation demonstrated that doses of aspirin ≤100 mg/d did not appear to increase postoperative bleeding. However, only 2 eligible studies were included in our study, thus more studies should be pooled to further verify the conclusion. The dose of 325 mg/d may not be a cutoff value that has clinical and statistical significance. Furthermore, sensitivity analyses were performed excluding 3 studies that reported aspirin doses more than 2000 mg/d for the RCTs and observational studies, respectively, because such high doses would be very likely to increase postoperative bleeding and are no longer used in clinical practice for antiplatelet therapy. We showed that despite the differences in effect sizes, the results were consistent before and after excluding the study of Fuller et al for the RCTs and that of Michelson et al and Torosian et al for the observational studies, respectively, suggesting the robust stability of our initial results. For efficacy outcomes, no difference in the rates of perioperative MI, CVAs, and mortality was statistically significant between the aspirin and the control groups. We noted publication bias for the outcome of MI in the RCTs. It is perplexing, and asymmetry among the RCTs was probably due to the absence of enough studies and the fact that Egger regression test has been challenged because of its high type I error rates (the proportion of false-positive results) when ORs are used.⁴⁶

Despite substantial advances in surgical technique in the recent years, complications of CABG continue to be troubling. Aimed at reducing the risks of MI and death from coronary artery disease, CABG correlates with its own risks of major adverse ischemic events (such as MI and CVAs).² Although its administration immediately before CABG may reduce thrombotic events and improve postoperative outcomes, the beneficial effects of preoperative aspirin may be counteracted by a competing risk of excessive intraoperative and postoperative bleeding.⁴⁷ Preoperative aspirin therapy for CABG is certainly one of the most important issues to be investigated in cardiovascular surgery because of disparate guidelines from professional organizations, conflicting practices of cardiac surgeons, and inconsistent results from the clinical trials.

It has long been established that the antiplatelet effects of aspirin are primarily mediated by its ability to irreversibly inhibit platelet cyclooxygenase 1 (COX-1) and cyclooxygenase 2 (COX-2), thereby impairing platelet aggregation.⁴⁷ Inhibition of COX-2 at higher aspirin doses may favor prothrombotic events, but it is 170-fold more potent in inhibiting COX-1 than COX-2.⁴⁸ Endothelial injuries of grafts resulting from operative procedures and utilization of cardiopulmonary bypass technology are believed to increase the risks of thrombotic events in CABG.^49,50 Within 1 month after surgery, 3% to 12% of saphenous vein grafts occlude, which is associated with 9% of mortality.^49,51 Early postoperative aspirin has been shown to, short-term and long-term, prevent graft occlusion, reduce ischemic events, and improve survival.² There is class 1A evidence that early aspirin administration after CABG reduces thrombotic saphenous vein grafts occlusion as well as overall morbidity and mortality. The AHA/ACC and the European Association of Cardiothoracic Surgeons recommend that 150 to 325 mg aspirin be administered within 6 hours following CABG.³ In addition, aspirin continues to be effective in preventing vein-graft occlusion late after the operation.⁵² Preoperative aspirin therapy might also be beneficial for patients with perioperative CABG. Furthermore, recent evidence has suggested that sudden withdrawal of long-term antiplatelet therapy may initiate a prothrombotic state, leading to excess risks of ischemic events perioperatively.^8,53 This is the most common case for those awaiting elective CABG. However, it remains to be clearly determined whether the benefits of preoperative aspirin may outweigh the hazards of excessive bleeding attributable to its use prior to CABG. Low doses of aspirin have been suggested to be as effective as higher doses for antiplatelet therapy while minimizing the risk of bleeding. A daily dose of 100 mg is commonly recommended for primary prevention of coronary artery disease.¹ Our analysis demonstrated that a daily dose of ≤100 mg before CABG may not be associated with an excess of bleeding. Sun et al⁹demonstrated that doses < 325 mg/d before CABG did not appear to increase postoperative bleeding. However, the effects of low-dose preoperative aspirin on patients with CABG need to be further clarified. Our meta-analysis suggested that preoperative aspirin was associated with increased transfusions in RCTs and observational studies and increased reoperation for bleeding in RCTs. Moreover, excessive postoperative bleeding leads to postoperative hypovolemia and hypotension, postoperative tamponade, and higher rates of blood transfusion.^54
–56 However, it is not well established that an increase in postoperative bleeding of around 200 mL is clinically relevant because of a lack of solid evidence. Preoperative aspirin did not increase reoperation for bleeding in the observational studies, and moreover, there was heterogeneity that we failed to eliminate for the results of transfusions. Thus, the clinical relevance of increased postoperative drainage in the aspirin groups with regard to some significant outcomes remains to be further elucidated.

In the past 30 years, many great advances in CABG surgery have taken place, and their links to our end points should be taken into consideration. As we included historical studies conducted for more than 30 years, it was critical to distinguish those outdated studies of low quality. Hence, we carried out subgroup analyses by Jadad score (for RCTs only), date of publication, and sample size in order to not only explore the effects of these factors on heterogeneity but also pool those relatively high-quality studies to obtain more robust estimates of our end points. A 2009 study by Gulbins et al¹⁵ reviewed a sum of more than 10 000 patients with CABG from 1995 to 2007 and showed that preoperative aspirin therapy did not seem to influence the postoperative bleeding. The authors did not provide us with detailed information regarding aspirin doses. All patients received antifibrinolytic therapy using aprotinin, which might explain why postoperative bleeding did not differ between the 2 groups. In our analysis, postoperative bleeding was shown to be increased in the aspirin groups in both early and late era, but the effect size of the latter was smaller. The use of blood conservation strategies and improvements in cardiopulmonary bypass technology, which lessens the damage to cellular components, may account for such a difference.¹⁰ In addition, we failed to resolve the problem of heterogeneity when we analyzed the results of transfusion requirements. The lack of homogeneity could be attributed to interhospital differences in the use of blood products and adherence to transfusion protocols.^57,58 Moreover, Dacey et al¹³ and Bybee et al¹² showed lower risks of fatal outcomes or mortality in the preoperative aspirin groups. Mikkola et al¹⁴ demonstrated that preoperative aspirin lowered the risk of postoperative stroke. However, our analysis had insufficient power to evaluate the effects of preoperative aspirin on the efficacy end points.

Inhibition of platelet activity by aspirin is not always achieved, and recent evidence has noted a wide variability in the response to aspirin. Residual platelet reactivity or aspirin resistance, measured by platelet function tests after aspirin administration, might be an important risk factor for adverse ischemic events in CABG.⁵⁹ Inversely, a pronounced response to aspirin potentially places perioperative patients at an excess risk of bleeding.⁵⁵ Although still in the early stage of research and development, individually tailored dose-dependent aspirin therapy based on platelet function tests is likely to be clinically employed in the near future.⁶⁰ Since the reported incidence of aspirin resistance varied widely in the literature, the lack of platelet response to aspirin quantification may constitute a major drawback of current RCTs and observational studies, and it is critical to adjust for this variable in the future trials. The duration of aspirin effects is also individually variable. Therefore, most studies had inherently diluted intergroup comparisons and reduced capacity to define safety and efficacy outcomes of preoperative aspirin when dichotomizing their participants into early and late discontinuation groups using a cutoff value of 3, 5, or 7 days.⁶¹ Another confounder we wish to point out is the administration of other antiplatelet or antifibrinolytic drugs in patients with CABG. Intraoperative use of antifibrinolytics has been revealed to significantly reduce postoperative blood loss in cardiac surgeries, and it might lower or even eliminate the bleeding risk of preoperative aspirin.⁶² Dual antiplatelet therapy is commonly prescribed for a portion of patients awaiting CABG, especially ones receiving percutaneous coronary intervention before surgery.⁶³ Thus, inclusion of studies either using antifibrinolytic drugs or not reporting detailed data for use of drugs rather than aspirin possibly has the potential for confounding.

In 2012, Deja et al²⁹ reported a single-center, double-blind RCT involving patients undergoing first-time elective CABG. The patients were assigned to receive either 300 mg of aspirin (n = 387) or placebo (n = 396) the night before the coronary bypass surgery. The trial revealed that performing coronary grafts on aspirin is associated with increased postoperative bleeding but may decrease the long-term hazard of coronary events. Our analysis was consistent with the results from Deja et al with regard to postoperative bleeding. However, high-level evidence regarding preoperative aspirin administration in patients with CBAG is poorly sourced and currently limited to observational studies with a few RCTs. A lack of high-quality prospective studies may account for the divergent opinions on preoperative aspirin in CABG at present. Sun et al and Alghamdi et al called for high-quality prospective trials to assess the association between preoperative aspirin therapy and important postoperative outcomes in the settings of contemporary cardiac surgical practice.^9,10

Limitations

There are several limitations to our meta-analysis. First, we pooled the results of RCTs and observational studies in the analysis of CVAs and MI. Although there has been controversy regarding pooling the results of different study designs, we noted no significant difference caused by study design. Second, inclusion of observational studies inevitably introduced a source of potential bias that inherently exists in nonrandomized, unblinded design. Third, differences between treatment groups in the doses of preoperative aspirin, use of postoperative aspirin, perioperative antifibrinolytics, and other antiplatelet drugs have been several confounders. Fourth, individual variability in response to aspirin and the duration of aspirin may dilute any effects of preoperative aspirin on our end points. Fifth, we failed to eliminate the heterogeneity in the analysis of transfusion requirements. Sixth, included studies were underpowered to determine significant difference in our efficacy outcomes. Finally, our studies did not have an analysis of only on-pump CABG, thus the risk benefit is unclear for this group.

Conclusions

Preoperative aspirin therapy increases postoperative bleeding, PRBC transfusion requirements, and reoperation for bleeding in patients undergoing CABG, which may be resolved by doses ≤100 mg/d. High-quality RCTs are required to obtain conclusive results for the effects of preoperative aspirin therapy on the safety and efficacy end points in patients with CABG.

Footnotes

Authors’ Note

The study was performed at the School of Medicine, Shandong University.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ. 2002;324(7329):71–86.

Mangano

. Aspirin and mortality from coronary bypass surgery. N Engl J Med. 2002;347(17):1309–1317.

Hillis

Smith

Anderson

. Special articles: 2011 ACCF/AHA guideline for coronary artery bypass graft surgery: executive summary: a report of the American college of cardiology foundation/American heart association task force on practice guidelines. Anesth Analg. 2012;114(1):11–45.

Sethi

Copeland

Goldman

Moritz

Zadina

Henderson

. Implications of preoperative administration of aspirin in patients undergoing coronary artery bypass grafting. department of veterans affairs cooperative study on antiplatelet therapy. J Am Coll Cardiol. 1990;15(1):15–20.

ElBardissi

Aranki

Sheng

O'Brien

Greenberg

Gammie

. Trends in isolated coronary artery bypass grafting: an analysis of the society of thoracic surgeons adult cardiac surgery database. J Thorac Cardiovasc Surg. 2012;143(2):273–281.

Ferraris

Moliterno

. The society of thoracic surgeons practice guideline series: aspirin and other antiplatelet agents during operative coronary revascularization (executive summary). Ann Thorac Surg. 2005;79(4):1454–1461.

Eagle

Guyton

Davidoff

. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: summary article. a report of the American college of cardiology/American heart association task force on practice guidelines (committee to update the 1999 guidelines for coronary artery bypass graft surgery). J Am Coll Cardiol. 2004;44(5):e213–e310.

Lordkipanidze

Diodati

Pharand

. Possibility of a rebound phenomenon following antiplatelet therapy withdrawal: a look at the clinical and pharmacological evidence. Pharmacol Therapeut. 2009;123(2):178–186.

Sun

Whitlock

Cheng

. 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(8):1057–1071.

10.

Alghamdi

Moussa

Fremes

. Does the use of preoperative aspirin increase the risk of bleeding in patients undergoing coronary artery bypass grafting surgery? Systematic review and meta-analysis. J Cardiac Surg. 2007;22(3):247–256.

11.

Goldman

Copeland

Moritz

. Improvement in early saphenous vein graft patency after coronary artery bypass surgery with antiplatelet therapy: results of a veterans administration cooperative study. Circulation. 1988;77(6):1324–1332.

12.

Bybee

Powell

Valeti

. Preoperative aspirin therapy is associated with improved postoperative outcomes in patients undergoing coronary artery bypass grafting. Circulation. 2005;112(9 suppl):I286–I292.

13.

Dacey

Munoz

Johnson

. Effect of preoperative aspirin use on mortality in coronary artery bypass grafting patients. Ann Thorac Surg. 2000;70(6):1986–1990.

14.

Mikkola

Wistbacka

Gunn

. Timing of preoperative aspirin discontinuation and outcome after elective coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth. 2012;26(2):245–250.

15.

Gulbins

Malkoc

Ennker

. Preoperative platelet inhibition with ASA does not influence postoperative blood loss following coronary artery bypass grafting. Thorac Cardiovasc Surg. 2009;57(1):18–21.

16.

http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.

17.

Jadad

Moore

Carroll

. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17(1):1–12.

18.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–188.

19.

Lim

Ali

. Indirect comparison meta-analysis of aspirin therapy after coronary surgery. BMJ. 2003;327(7427):1309.

20.

Friedrich

Adhikari

Beyene

. Inclusion of zero total event trials in meta-analyses maintains analytic consistency and incorporates all available data. BMC Med Res Methodol. 2007;7:5.

21.

Higgins

JPT

Green

. Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 [Updated September 2006]. The Cochrane Library. Chichester: John Wiley & Sons, Ltd; 2006: issue 4.

22.

Fuller

JK CJ

. Does short-term pre-operative aspirin in coronary bypass patients increase post-operative bleeding? Vasc Surg. 1985;19(3):174–178.

23.

Karwande

Weksler

Gay

Jr Subramanian

. Effect of preoperative antiplatelet drugs on vascular prostacyclin synthesis. Ann Thorac Surg. 1987;43(3):318–322.

24.

Ferraris

Lough

Berry

. Preoperative aspirin ingestion increases operative blood loss after coronary artery bypass grafting. Ann Thorac Surg. 1988;45(1):71–74.

25.

Goldman

Copeland

Moritz

. Starting aspirin therapy after operation. effects on early graft patency. Department of veterans affairs cooperative study group. Circulation. 1991;84(2):520–526.

26.

Hockings

Ireland

Gotch-Martin

Taylor

. Placebo-controlled trial of enteric coated aspirin in coronary bypass graft patients. Effect on graft patency. Med J Aust. 1993;159(6):376–378.

27.

Kallis

Tooze

Talbot

Cowans

Bevan

Treasure

. Pre-operative aspirin decreases platelet aggregation and increases post-operative blood loss—a prospective, randomised, placebo controlled, double-blind clinical trial in 100 patients with chronic stable angina. Eur J Cardiothorac Surg. 1994;8(8):404–409.

28.

Morawski

Sanak

Cisowski

. Prediction of the excessive perioperative bleeding in patients undergoing coronary artery bypass grafting: role of aspirin and platelet glycoprotein IIIa polymorphism. J Thorac Cardiovasc Surg. 2005;130(3):791–796.

29.

Deja

Kargul

Domaradzki

. Effects of preoperative aspirin in coronary artery bypass grafting: a double-blind, placebo-controlled, randomized trial. J Thorac Cardiovasc Surg. 2012;144(1):204–209.

30.

Michelson

Morganroth

Torosian

Mac Vaugh

H III

. Relation of preoperative use of aspirin to increased mediastinal blood loss after coronary artery bypass graft surgery. J Thorac Cardiovas Surg. 1978;76(5):694–697.

31.

Taggart

Siddiqui

Wheatley

. Low-dose preoperative aspirin therapy, postoperative blood loss, and transfusion requirements. Ann Thorac Surg. 1990;50(3):424–428.

32.

Torosian

Michelson

Morganroth

MacVaugh

H III

. Aspirin- and coumadin-related bleeding after coronary- artery bypass graft surgery. Ann Intern Med. 1978;89(3):325–328.

33.

Rawitscher

Jones

McCoy

Lindsley

. A prospective study of aspirin's effect on red blood cell loss in cardiac surgery. J Cardiovas Surg. 1991;32(1):1–7.

34.

Reich

DL PG

Vela-Cantos

Bodian

Lansman

. Aspirin does not increase homologous blood requirements in elective coronary bypass surgery. Anesth Analg. 1994;79(1):4–8.

35.

Vuylsteke

Oduro

Cardan

Latimer

. Effect of aspirin in coronary artery bypass grafting. J Cardiothorac Vasc Anesth. 1997;11(7):831–834.

36.

Jakics

Lee

Ikeda

. Preoperative aspirin and heparin therapy does not increase perioperative blood loss and blood product requirements in coronary artery bypass graft surgery. J Anesth. 1999;13(1):8–13.

37.

Weightman

Gibbs

Weidmann

. The effect of preoperative aspirin-free interval on red blood cell transfusion requirements in cardiac surgical patients. J Cardiothorac Vasc Anesth. 2002;16(1):54–58.

38.

Ray

Deniz

Olivieri

. Increased blood product use among coronary artery bypass patients prescribed preoperative aspirin and clopidogrel. BMC Cardiovasc Disord. 2003;3:3.

39.

Gerrah

Izhar

. Beneficial effect of aspirin on renal function post-cardiopulmonary bypass. Asian Cardiovasc Thorac Ann. 2003;11(4):304–308.

40.

Gerrah

Ehrlich

Tshori

Sahar

. Beneficial effect of aspirin on renal function in patients with renal insufficiency postcardiac surgery. J Cardiovasc Surg. 2004;45(6):545–550.

41.

Gerrah

Elami

Stamler

Smirnov

Stoeger

. Preoperative aspirin administration improves oxygenation in patients undergoing coronary artery bypass grafting. Chest. 2005;127(5):1622–1626.

42.

Srinivasan

Grayson

Pullan

Fabri

Dihmis

. Effect of preoperative aspirin use in off-pump coronary artery bypass operations. Ann Thorac Surg. 2003;76(1):41–45.

43.

Kamran

Ahmed

Dar

Khan

. Effect of aspirin on postoperative bleeding in coronary artery bypass grafting. Ann Thorac Cardiovasc Surg. 2008;14(4):224–229.

44.

Hijazi

. Aspirin does not increase bleeding and allogeneic blood transfusion in coronary artery surgery. Thorac Cardiovasc Surg. 2011;59(7):421–424.

45.

Jacob

Smedira

Blackstone

Williams

Cho

. Effect of timing of chronic preoperative aspirin discontinuation on morbidity and mortality in coronary artery bypass surgery. Circulation. 2011;123(6):577–583.

46.

Peters

Sutton

Jones

Abrams

Rushton

. Comparison of two methods to detect publication bias in meta-analysis. JAMA. 2006;295(6):676–680.

47.

Kulik

Chan

Ruel

. Antiplatelet therapy and coronary artery bypass graft surgery: perioperative safety and efficacy. Expert Opin Drug Saf. 2009;8(2):169–182.

48.

Altman

Luciardi

Muntaner

Herrera

. The antithrombotic profile of aspirin. aspirin resistance, or simply failure? Thromb J. 2004;2(1):1.

49.

Motwani

Topol

. Aortocoronary saphenous vein graft disease: pathogenesis, predisposition, and prevention. Circulation. 1998;97(9):916–931.

50.

Levy

Tanaka

. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg. 2003;75(2):S715–S7120.

51.

Fabricius

Gerber

Hanke

Garbade

Autschbach

Mohr

. Early angiographic control of perioperative ischemia after coronary artery bypass grafting. Eur J Cardiothorac Surg. 2001;19(6):853–858.

52.

Chesebro

Fuster

Elveback

. Effect of dipyridamole and aspirin on late vein-graft patency after coronary bypass operations. N Engl J Med. 1984;310(4):209–214.

53.

Beving

Zhao

Albage

Ivert

. Abnormally high platelet activity after discontinuation of acetylsalicylic acid treatment. Blood Coagul Fibrinolysis. 1996;7(1):80–84.

54.

Ranucci

Bozzetti

Ditta

Cotza

Carboni

Ballotta

. Surgical reexploration after cardiac operations: why a worse outcome? Ann Thorac Surg. 2008;86(5):1557–1562.

55.

Ferraris

Joseph

Wehner

Mentzer

Jr . Aspirin and postoperative bleeding after coronary artery bypass grafting. Ann Surg. 2002;235(6):820–827.

56.

Mehta

Sheng

O'Brien

. Reoperation for bleeding in patients undergoing coronary artery bypass surgery: incidence, risk factors, time trends, and outcomes. Circ Cardiovasc Qual Outcomes. 2009;2(6):583–590.

57.

Surgenor

Churchill

Wallace

. The specific hospital significantly affects red cell and component transfusion practice in coronary artery bypass graft surgery: a study of five hospitals. Transfusion. 1998;38(2):122–134.

58.

Stover

Siegel

Parks

. Variability in transfusion practice for coronary artery bypass surgery persists despite national consensus guidelines: a 24-institution study. Institutions of the multicenter study of perioperative ischemia research group. Anesthesiology. 1998;88(2):327–333.

59.

Ben-Dor

Kleiman

Lev

. Assessment, mechanisms, and clinical implication of variability in platelet response to aspirin and clopidogrel therapy. Am J Cardiol. 2009;104(2):227–233.

60.

Petricevic

Biocina

Konosic

Ivancan

. Timing of preoperative aspirin discontinuation and outcome after elective coronary artery bypass graft surgery: a role of platelet function assessment. J Cardiothorac Vasc Anesth. 2012;26(4):e35–e36

61.

Myles

. Stopping aspirin before coronary artery surgery: between the devil and the deep blue sea. Circulation. 2011;123(6):571–573.

62.

Henry

Carless

Moxey

. Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev. 2007;(4):CD001886.

63.

Kopjar

Biocina

Petricevic

. Variable on treatment platelet reactivity in coronary artery bypass grafting patients suggests the need for perioperative platelet function testing. J Thorac Cardiovasc Surg. 2012;144(5):1274–1275; author reply 5-6.

Safety and Efficacy Outcomes of Preoperative Aspirin in Patients Undergoing Coronary Artery Bypass Grafting

Abstract

Background:

Methods:

Results:

Conclusions:

Keywords

Introduction

Methods

Eligibility Criteria

Search Strategy

Data Extraction

Data Analysis

Results

Eligible Studies

Study Characteristics

Safety Outcomes

Postoperative Bleeding

Randomized controlled trials

Observational studies

Packed Red Blood Cell Transfusion Requirements

Reoperation for Bleeding

Randomized controlled trials

Observational studies

Efficacy Outcomes

Myocardial Infarction

Cerebrovascular Events

Mortality

Publication Bias Analysis

Discussion

Limitations

Conclusions

Footnotes

Authors’ Note

Declaration of Conflicting Interests

Funding

References