ORIGINAL ARTICLE
Association Between Preoperative Aspirin-dosing Strategy and Mortality After Coronary Artery Bypass Graft Surgery Yi Deng, MD, Paul V. Pisklak, MD, Vei-Vei Lee, MS,y Daniel A. Tolpin, MD, z Charles D. Collard, MD, z MacArthur A. Elayda, MD, PhD,y Joseph Coselli, MD,§ and Wei Pan, MD z
Objective: To determine whether preoperative aspirin—acetylsalicylic acid (ASA)—timing or dose independently affects 30-day all-cause mortality. Background: Preoperative ASA administration is associated with reduced morbidity and mortality after coronary artery bypass graft (CABG). However, data are lacking regarding optimal timing and dosing of ASA. Methods: We retrospectively reviewed data from 3018 consecutive patients who underwent CABG surgery between July 2005 and May 2011. Patients were assigned to 3 groups according to the time of the last preoperative ASA dose: (1) 24 hours or less preoperatively (n ¼ 1173), (2) between 24 and 72 hours (n ¼ 876), and (3) more than 72 hours or none (n ¼ 969). In a separate analysis, patients were grouped according to ASA dose: 81 mg (n ¼ 1285), 325 mg (n ¼ 1004), and none (n ¼ 543). The primary outcome was 30-day allcause mortality. Results: The 30-day mortality rate was significantly lower in patients who took ASA 24 hours or less preoperatively (1.5%) than in those who took it between 24 and 72 hours (3.2%) or more than 72 hours or none (2.9%). Multivariate analysis showed that ASA within 24 hours preoperatively was associated with reduced mortality (odds ratio [OR], 0.41; 95% confidence interval [CI], 0.20–0.82; P ¼ 0.01). Moreover, mortality was significantly reduced for patients taking 81 mg of ASA (1.4%) compared with 325 mg (2.9%) or none (3.9%). Multivariate analysis demonstrated that 81 mg of ASA decreased mortality risk by 66% (OR, 0.34; 95% CI, 0.18–0.66; P < 0.01), whereas 325 mg of ASA had no mortality benefit (OR, 0.74; 95% CI, 0.41– 1.35; P ¼ 0.33) compared with no ASA. Conclusions: Low-dose ASA use within 24 hours of CABG is independently associated with decreased early postoperative mortality.
bleeding risk have examined the continued use of ASA up to the time of surgery; in most studies, ASA was stopped 5 to 7 days preoperatively.12 Prematurely discontinuing ASA can lead to a prothrombotic state in patients who are already at higher risk of ischemia to vital organs.13 Two retrospective studies demonstrated that preoperative ASA use, despite showing a trend toward increased bleeding and transfusion, improved surgical outcomes including mortality.14,15 Furthermore, a large prospective study demonstrated that stopping ASA before CABG was an independent risk factor for increased mortality.5 Because the bleeding risk can vary significantly depending on the time and dose of last taken, current guidelines cite a wide range (from 3 to 10 days) for the optimal time to discontinue ASA before cardiac surgery.16,17 Furthermore, most studies of preoperative ASA in CABG surgery were not powered to examine important perioperative outcomes such as mortality, and they varied widely in ASA dose. To our knowledge, there is no published comparative study examining the effect of the interval of ASA discontinuation or the dose taken before elective CABG on 30-day all-cause mortality. Thus, the aim of this study was to determine whether the time of discontinuation or the dose of preoperative ASA was independently associated with a reduced incidence of 30-day all-cause mortality after primary on-pump CABG. To this end, we performed a multivariate stepwise logistic regression analysis on data from more than 3000 consecutive patients undergoing primary CABG at our institution, controlling for patient demographics, medical history, and preoperative medication.
Keywords: aspirin, coronary artery bypass graft, mortality
METHODS
(Ann Surg 2015;262:1150–1156)
A
spirin—acetylsalicylic acid (ASA)—has become a mainstay treatment in both primary and secondary prevention of patients with cardiovascular disease.1,2 In coronary artery bypass grafting (CABG), early postoperative ASA increases graft patency, reduces ischemic complications, improves survival, and is recommended within hours of surgery.3 –5 However, preoperative ASA use remains controversial. Many surgeons recommend patients discontinue ASA 7 days before CABG for fear of excessive bleeding, which can lead to increased transfusion requirement, greater complications, and increased morbidity and mortality.6 –11 However, evidence in the literature is conflicting, and few studies of preoperative ASA and From the Department of Anesthesiology, Baylor College of Medicine, Houston, TX; yDepartment of Biostatistics and Epidemiology, Texas Heart Institute, Houston, TX; zDivision of Cardiovascular Anesthesiology, Texas Heart Institute, Baylor St. Luke’s Medical Center, Houston, TX; and §Department of Cardiovascular Surgery, Texas Heart Institute, Houston, TX. Disclosure: The authors declare no conflicts of interest. Reprints: Wei Pan, MD, Division of Cardiovascular Anesthesiology, Texas Heart Institute, 6720 Bertner Ave, Houston, TX 77030. E-mail: [email protected]. Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0003-4932/14/26105-0821 DOI: 10.1097/SLA.0000000000000951
1150 | www.annalsofsurgery.com
Study Design In this retrospective cohort study, we reviewed data from consecutive patients undergoing primary isolated CABG surgery with cardiopulmonary bypass (n ¼ 3018) between July 2005 and May 2011 at the Texas Heart Institute (THI). We grouped patients according to the interval between their last dose of ASA and the time of surgery: (1) those who took ASA 24 hours or less before surgery (n ¼ 1173), (2) those who took ASA between 24 and 72 hours before surgery (n ¼ 876), and (3) those who took ASA more than 72 hours before surgery or did not take it at all (n ¼ 969). In a separate analysis, patients were also analyzed on the basis of ASA dose: 81 mg (n ¼ 1285), 325 mg (n ¼ 1004), and none (n ¼ 543). Because of their limited number, patients taking 162 mg of ASA (n ¼ 35) were excluded from the analysis. We also excluded patients who underwent concomitant valve or other cardiac surgery (eg, septal defect repair and ventricular aneurysm resection). The study was approved by the institutional review board at Baylor St. Luke’s Medical Center.
Data Collection Patient demographics, preoperative risk factors and medications, and the incidence of adverse postoperative outcomes were obtained from the cardiac surgical database at the THI. The timing Annals of Surgery Volume 262, Number 6, December 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Preoperative ASA in CABG and Mortality
and dosage of ASA were gathered from patients’ chart when available. For those who were admitted to the hospital on the day of surgery, the dose and time of the last ASA taken were obtained from a preoperative medication questionnaire that was completed as part of the standard preadmission work-up for each patient. For those who were already inpatients, the nursing medication administration record was used to gather the time and dose of the last ASA before surgery. The measured adverse outcomes included 30-day all-cause mortality, myocardial infarction (MI), stroke, and ventilator dependency. A diagnosis of MI was made if there were new Q waves, new persistent ST-segment or T-wave changes, or evidence of acute MI on autopsy. A diagnosis of stroke was made if there was clinical evidence or evidence of a focal or global defect on computed tomography, magnetic resonance imaging, or autopsy. Ventilator dependency was defined as mechanical ventilation for more than 48 hours. Although we assessed the incidence of reoperation for bleeding, the quantitative blood loss during initial surgery or the units and types of transfused blood products were not available in the database.
associated with a reduction in the risk of 30-day all-cause mortality after primary CABG, we performed a multivariate, stepwise, logistic regression after adjusting for potentially confounding variables. Odds ratios and corresponding 95% confidence intervals (CIs) are reported with the associated P values. To further control for selection bias, we used 27 perioperative variables to create propensity-matched populations for 2 separate analyses. In the first analysis, patients receiving ASA 24 hours or less before surgery were compared with all others, and in the second analysis, patients receiving 81 mg of ASA were compared with all others. Propensity scores were calculated for each patient, and patients were matched one-to-one between the 2 groups. After successful matching, we performed a multivariate, stepwise logistic regression to determine whether ASA timing and dose were independently associated with a reduction in the risk of 30-day all-cause mortality after primary CABG surgery.
Statistical Analysis All statistical analyses were performed using SAS statistical software (v9.1, SAS Institute, Inc, Cary, NC) by the Division of Biostatistics and Epidemiology at the THI. Categorical (frequency) variables are expressed as percentages, and continuous variables are expressed as the mean standard deviation. We performed a univariate analysis using the x2 test for categorical data and a t test for continuous variables to compare preoperative patient characteristics and medication, perioperative risk factors, and perioperative outcomes, with the primary endpoint being 30-day all-cause mortality. To determine whether ASA timing or dose was independently
RESULTS Between July 2005 and May 2011, 3018 patients underwent elective primary CABG at the THI. Of the total group of patients, 39% took ASA 24 hours or less before surgery, 29% discontinued it between 24 to 72 hours, and 32% either stopped it more than 72 hours before surgery or never took it (Table 1). Overall, the 3 groups were similar with regard to age, the number of men, and the history of smoking, obesity, hypertension, peripheral vascular disease, previous stroke, renal insufficiency, and diabetes (Table 1). However, patients who had a previous MI, unstable angina at the time of admission, urgent surgery, or worse heart failure symptoms (New York Heart Association class III/IV) took ASA more frequently within 72 hours of surgery. Patients who continued ASA closer to surgery were less likely to be on an intra-aortic balloon pump. Concurrent b-blocker
TABLE 1. Perioperative Demographics and Risk Factors for Patients Grouped by ASA Timing Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass (%) Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Clopidogrel, % Previous IABP, % Cross-clamp time, min Total bypass time, min
24 hrs (n ¼ 1173)
24 to 72 hrs (n ¼ 876)
>72 hrs or None (n ¼ 969)
P
49.3 11.9 77.5 29.6 66.3 32.1 42.8 56.8 12.0 48.9 38.7 5.9 87.6 23.4 52.6 28.5 86.9 43.7 6.8 35.0 21.6 59.7 52.3 27.0 5.8 40.0 66.2
48.9 11.8 77.4 29.4 67.2 33.6 44.6 60.3 12.9 50.5 41.3 8.0 87.6 23.0 52.6 27.7 85.1 47.0 8.5 34.6 24.0 56.6 50.1 25.7 6.1 41.2 68.2
50.2 10.2 73.5 29.2 63.8 30.0 35.9 52.7 10.0 37.0 30.8 7.4 88.0 24.5 52.1 26.8 83.9 45.5 8.0 32.9 22.8 55.6 49.2 30.3 6.9 38.9 65.2
NS NS NS NS NS NS < 0.01 < 0.01 NS < 0.01 < 0.01 NS NS NS NS NS NS NS NS NS NS NS NS NS < 0.01 NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intra-aortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1151
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Deng et al
30-Day All-Cause Mortality (%)
15.0
10.0
5.0
n = 876 n = 1173*
n = 969
3.2
2.9
1.5 0.0
≤ 24 hours
> 24 hours - ≤ 72 hours > 72 hours + none ASA Taken in Hours Prior to Surgery
*P < 0.05 in multivariate analysis performed for 30-day mortality.
FIGURE 1. Incidence of 30-day mortality in patients undergoing CABG on the basis of the time of discontinuation of aspirin before surgery. P < 0.05 in multivariate analysis performed for 30-day mortality. and angiotensin-converting enzyme (ACE) inhibitor use did not differ among the groups. There were no significant differences in ASA use before 2006, from 2007 to 2008, and after 2009 (P ¼ 0.20; data not shown). When stratified by the time of discontinuation of ASA before surgery, the incidence of 30-day all-cause mortality was 1.5% in patients who stopped ASA 24 hours or less before surgery, 3.2% in those who stopped between 24 and 72 hours, and 2.9% in those who stopped more than 72 hours before surgery or never took ASA (Fig. 1). In a multivariate analysis controlling for patient demographics, preoperative risk factors, and intraoperative variables, we found that taking ASA within 24 hours of surgery was independently associated with reduced mortality [odds ratio (OR), 0.41; 95% CI, 0.20–0.82; P ¼ 0.01]. There was no significant difference between the groups in other outcomes such as incidence of MI, stroke, ventilator dependency, or reoperation for bleeding (Table 2). In addition, the 30-day all-cause mortality of surgery was 2.5% in 2005 to 2006, 2.4% in 2007 to 2008, and 2.4% in 2009 to 2011. There were no significant differences in the death rate among these years (P ¼ 0.98). To assess the effect of preoperative ASA dose on mortality, we separated the patients into 3 groups: low dose (81 mg), high dose (325 mg), or no ASA (Table 3). Overall, 45% of patients were on low dose, 35% on high dose, and 19% did not take ASA. Predictably, patients who had a previous MI, urgent surgery, unstable angina, and worse heart failure symptoms were more likely to be taking ASA preoperatively, especially the dose of 325 mg. Moreover, the incidence of hyperlipidemia was greater in patients receiving preoperative ASA than in those receiving no ASA. However, the incidence of hypertension, stroke, diabetes, or smoking was similar between groups. Both groups of ASA takers were also more likely to be on b-blockers and ACE inhibitors at the same time. Interestingly,
older patients (age >65 years) were less likely to be on high-dose ASA than on low-dose or no ASA. Figure 2 shows the incidence of 30-day all-cause mortality according to ASA dose. Mortality was 1.4% for patients taking 81 mg of ASA, 2.9% for those taking 325 mg of ASA, and 3.9% for those not taking ASA (P < 0.01). Multivariate analysis showed that mortality was 66% lower in patients on low-dose ASA than in those not receiving preoperative ASA (OR, 0.34; 95% CI, 0.18–0.66; P < 0.01), whereas no mortality benefit was seen with high-dose ASA (OR, 0.74; 95% CI, 0.41–1.35; P ¼ 0.33). The incidence of postoperative MI and stroke did not differ among all 3 groups (Table 4). Preoperative ASA did not result in higher rates of reoperation for bleeding compared with no ASA, and the rates of reoperation were similar for the 325-mg and 81mg groups (3.8% vs 4.1%, respectively). To further control for selection bias related to the choice of therapy, we used 27 variables to create propensity-matched populations for 2 separate analyses. In the first cohort, we successfully matched patients on a one-to-one basis who discontinued ASA use within 24 hours of surgery (n ¼ 1173) with those who discontinued it more than 24 hours after surgery (n ¼ 1173) (Table 5). In the second cohort, patients who took 81 mg of ASA preoperatively were matched with all other patients who were not taking low-dose ASA (n ¼ 1053 for each group) (Table 6). We then performed stepwise logistic regression on the matched cohorts. Even after controlling for propensity scores, the observed reduction in 30day all-cause mortality associated with low-dose ASA use persisted in the multivariate analysis. Other significant predictors of early mortality included older age (OR, 2.3; 95% CI, 1.2–4.5; P ¼ 0.03), previous MI (OR, 2.7; 95% CI, 1.4–5.0; P < 0.01), and preoperative renal dysfunction (OR, 4.2; 95% CI, 2.2–7.7; P < 0.01) (Fig. 3).
DISCUSSION In this single-center study of preoperative ASA use in 3018 patients undergoing primary isolated CABG, we found that taking ASA within 24 hours of surgery was associated with a 48% relative reduction in 30-day all-cause mortality. This finding held true even when we adjusted for confounding risk factors, severity of illness, and other medication use. In a separate analysis, we found that preoperative low-dose (81 mg) ASA was independently associated with a reduced 30-day mortality rate when compared with no ASA, but preoperative high-dose (325 mg) ASA did not confer a mortality benefit. These data suggest that preoperative low-dose ASA therapy within 24 hours of surgery would have a favorable effect on patients undergoing primary CABG surgery. ASA therapy has been shown to reduce mortality in the setting of acute coronary syndrome.18,19 Although the mechanism underlying the association of preoperative ASA and reduced surgical mortality is unknown, increased platelet activation during CABG surgery can lead to thrombotic or ischemic events. Thus, it is reasonable to postulate that ASA can confer a mortality benefit by limiting platelet activation, thereby reducing graft thrombosis as well as other cerebrovascular, gastrointestinal, or renal
TABLE 2. Incidence of Adverse Postoperative Outcomes as Grouped by ASA Timing 24 hrs (n ¼ 1173)
24 to 72 hrs (n ¼ 876)
>72 hrs or None (n ¼ 969)
P
1.5 3.1 3.9 2.5
3.2 4.3 5.4 2.6
2.9 3.9 3.4 2.6
0.02 NS NS NS
30-d mortality, % MI, % Reoperation for bleeding, % Stroke, % NS indicates not significant.
1152 | www.annalsofsurgery.com
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Preoperative ASA in CABG and Mortality
TABLE 3. Perioperative Demographics and Risk Factors for Patients Grouped by ASA Dose Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Clopidogrel, % Previous IABP, % Cross-clamp time, min Total bypass time, min
None (n ¼ 543)
81 mg (n ¼ 1285)
325 mg (n ¼ 1004)
P
50.1 10.7 72.4 29.6 65.7 31.8 39.0 42.1 10.1 42.1 29.1 8.1 85.4 22.1 50.3 28.1 81.0 45.0 7.4 34.1 21.7 52.5 45.4 21.5 5.7 39.6 67.4
52.6 11.2 77.5 29.2 63.9 30.2 37.9 42.7 11.5 42.7 34.0 7.6 88.8 24.1 52.2 25.9 87.9 44.6 6.8 31.6 23.3 60.6 53.6 32.8 4.4 39.8 65.6
45.0 12.0 77.3 29.7 68.3 33.9 46.8 51.7 13.0 51.7 46.4 5.5 88.0 24.0 54.3 29.7 85.3 46.4 9.1 36.3 22.5 57.1 50.9 25.7 6.2 40.6 66.8
< 0.01 NS 0.02 NS NS NS < 0.01 0.03 NS < 0.01 < 0.01 NS NS NS NS NS < 0.01 NS NS NS NS < 0.01 < 0.01 < 0.01 NS NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intra-aortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
complications.5,20–23 Our study showed a decrease in early mortality but not a reduction in postoperative MI or stroke. These results are similar to those of other outcome studies of preoperative ASA in CABG surgery, suggesting that ASA can reduce post-CABG mortality in ways other than decreasing cardiac or cerebrovascular complications.14,24 Because our institution did not routinely examine postoperative cardiac enzyme levels, our ability to detect postoperative MI or cardiac dysfunction was limited. Thus, we focused on the well-defined clinical endpoint of early all-cause mortality as our primary outcome. In our study, preoperative low-dose ASA continued up to the time of surgery significantly decreased short-term postsurgical
30-Day All-Cause Mortality (%)
15.0
10.0
5.0
n = 543 n = 1004
3.9
n = 1285*
2.9
TABLE 4. Incidence of Adverse Postoperative Outcomes as Grouped by ASA Dose
1.4
0.0
None
81 mg ASA Dose Prior to Surgery
325 mg
*P < 0.05 in multivariate analysis performed for 30-day mortality.
FIGURE 2. Incidence of 30-day mortality in patients undergoing CABG on the basis of aspirin dose before surgery. P < 0.05 in multivariate analysis performed for 30-day mortality. ß
mortality. To our knowledge, this is the only study that specifically compared intervals of ASA use as close as 24 hours within surgery. In most comparative studies, ASA use is discontinued 7 days before surgery, which likely underestimates its potential benefits. The clinical effect of ASA is typically shorter than 7 days, and some studies suggest that platelet inhibition is minimal after 3 days and can vary significantly among individuals.25–27 Also, surgeons have been reluctant to continue ASA closer than 7 days before surgery for fear of increased bleeding, given that cardiopulmonary bypass can further decrease platelet function postoperatively through hypothermia, sequestration, and mechanical factors.28,29 This concern was supported by studies published in the 1980 s to early 1990 s that showed significantly increased bleeding and transfusion-related complications associated with preoperative ASA use in CABG surgery.12 However, over the last 2 decades, changes in clinical practices may have addressed some of these issues, tipping the scales in favor of preoperative ASA use. First, modern practices of aggressive bloodsalvaging techniques and the widespread adoption of antifibrinolytics such as e-aminocaproic acid may have blunted any additional
2015 Wolters Kluwer Health, Inc. All rights reserved.
30-d mortality, % MI, % Reoperation for bleeding, % Stroke, %
None (n ¼ 543)
81 mg (n ¼ 1285)
325 mg (n ¼ 1004)
P
3.9 4.8 4.8 2.7
1.4 3.1 4.1 2.4
2.9 3.7 3.8 2.6
< 0.01 NS NS NS
NS indicates not significant.
www.annalsofsurgery.com | 1153
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Deng et al
Annals of Surgery Volume 262, Number 6, December 2015
TABLE 5. Demographics of Propensity-matched Groups by the Time of ASA Discontinuation Before Surgery
TABLE 6. Demographics of Propensity-matched Groups by ASA Dose Before Surgery
Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % Pulmonary disease, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Statin use, % Previous IABP, % Cross-clamp time, min Total bypass time, min
24 hrs (n ¼ 1173)
>24 hrs (n ¼ 1173)
P
49.3 11.9 67.5 29.6 66.3 32.1 42.8 56.8 12.0 48.9 38.7 5.9 87.6 34.7 23.4 52.6 28.5 86.9 43.7 6.8 35.0 21.6 59.7 52.3 81.7 4.2 40.0 66.2
49.7 11.5 65.2 29.2 64.5 31.5 39.8 56.7 11.5 48.9 37.4 7.5 88.5 38.9 23.3 52.6 27.0 85.7 46.1 9.0 33.6 22.4 58.9 50.8 81.7 5.8 40.5 67.0
NS NS NS NS NS NS NS NS NS NS NS NS NS 0.04 NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % Pulmonary disease, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Statin use, % Previous IABP, % Cross-clamp time, min Total bypass time, min
81 mg (n ¼ 1053)
Not on 81 mg (n ¼ 1053)
P
48.6 11.9 78.2 29.3 67.5 32.1 40.2 57.0 12.1 45.5 36.6 7.1 88.7 36.6 24.0 52.3 27.0 87.4 44.3 4.8 34.2 22.6 59.2 53.9 81.1 5.2 40.7 67.5
48.6 11.2 74.8 29.6 68.1 31.2 39.7 57.3 11.7 45.0 37.1 6.4 88.7 35.3 23.1 53.0 29.2 85.7 47.7 5.4 35.0 23.7 58.8 51.6 81.7 5.1 40.2 66.6
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intraaortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intraaortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
bleeding effect from ASA. Moreover, recent studies have shown that stopping ASA within 3 days did not result in increased bleeding when compared with 7 days.11,30 This may enable patients to continue receiving the benefits of ASA therapy until surgery without the concern for increased bleeding complications.31,32 Second, when compared with the earlier studies, patients today are typically older, present with more comorbidities, are more likely to have coronary
stents, and are on chronic ASA therapy.5,12,31 ASA withdrawal before surgery can be associated with a 2% to 10% incidence of acute coronary syndrome and a 6% incidence of acute lower-limb ischemia.13 Thus, in today’s patients, the risk of ASA withdrawal may outweigh the risk of bleeding complications and may even adversely affect postoperative mortality. In fact, 2 large retrospective studies from this decade showed discontinuing ASA 5 to 7 days before
Independent Predictors
95% CI
PVD
1.2-4.0
Renal dysfunction
2.2-7.7
Previous MI
1.4-5.0
NYHA class III/IV
1.1-4.8
Female
1.1-3.0
Age > 65
1.2-4.5
ASA ≤ 24 hours
0.2-0.8
ASA 81 mg
0.2-0.7 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Odds Ratio (95% CI)
1154 | www.annalsofsurgery.com
ß
FIGURE 3. Multivariate analysis of independent predictors on 30-day all-cause mortality. PVD indicates peripheral vascular disease.
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
CABG was associated with increased postoperative mortality.14,15 Our study suggests that stopping ASA more than 24 hours before surgery may lead to worse outcomes. The data on the optimal dose of ASA before CABG surgery are limited because there are no trials that directly compare different dosing regimens and their outcomes. In preventing vein graft occlusions after CABG, low-dose ASA (75–325 mg) was not inferior to high-dose ASA (500–1500 mg).33 In interventional cardiology procedures for acute coronary syndrome, data demonstrate that higherdose ASA did not decrease the incidence of MI and all-cause mortality but did significantly increase bleeding complications.34 In a metaanalysis by Sun and colleagues12 examining the effect of preoperative ASA in CABG, the dose of ASA included in studies varied significantly between 80 and 2600 mg. They reported that the use of doses of 325 mg or more was associated with a significant increase in bleeding, whereas doses of less than 325 mg were not. The meta-analysis was not powered to examine the dose effect on mortality. Thus, it seems that given the equivocal benefit and the clearly identified risks of using high-dose ASA, low-dose ASA may be more beneficial for patients undergoing CABG. Our study demonstrated that low-dose (81 mg) ASA was independently associated with surgical mortality, whereas high-dose (325 mg) was not. It may be that the increased bleeding complications from high-dose ASA negated its benefits. However, we were unable to assess that hypothesis because of the lack of quantitative blood loss and transfusion-related data in our database. Our study has several limitations. First, this is a retrospective study, so we could not control for the administration of ASA preoperatively. Despite careful use of regression models to adjust for potential confounders, many factors may still affect the outcome. However, we attempted to limit bias by using propensity matching with a large number of variables to ensure the groups were as evenly matched as possible with regard to demographics and perioperative risk factors. Second, this study was conducted at a single institution; however, our perioperative outcomes were similar to recently published results from other large centers for cardiac surgery, suggesting that the experiences were comparable.14,15 Third, our database did not track the amount of bleeding or transfusion requirements, so we were unable to examine the relationship between preoperative ASA and increased bleeding complications. Nevertheless, the rates of reoperations for bleeding were similar in all groups and were comparable to recently published data, suggesting that the bleeding and transfusion requirements may not have differed significantly.14,15,30,31 Lastly, we did not measure or control for the effect of postoperative ASA administration, although standard practice at the THI is for all patients to begin ASA within 1 day after surgery.
CONCLUSIONS The timing and dose of ASA before CABG surgery are associated with the 30-day mortality rates. Specifically, ASA use (81 mg) within 24 hours of surgery is independently associated with decreased postoperative mortality. Future large-scale, randomized, controlled studies are warranted to further identify the role of preoperative ASA in perioperative outcome benefits in cardiac surgical patients.
ACKNOWLEDGMENTS We thank Rebecca Bartow, PhD, of the Texas Heart Institute, for editorial assistance.
REFERENCES 1. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk, patients. BMJ. 2002;324:71–86. 2. The´roux P, Ouimet H, McCans J, et al. Aspirin, heparin, or both to treat acute unstable angina. N Engl J Med. 1988;319:1105–1111.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Preoperative ASA in CABG and Mortality
3. Chesebro JH, Clements IP, Fuster V, et al. A platelet-inhibitor drug trial in coronary-artery bypass operations. N Engl J Med. 1982;307:73–78. 4. Goldman S, Copeland J, Moritz T, et al. Starting aspirin therapy after operation: effects on early graft patency. Department of Veterans Affairs Cooperative Study Group. Circulation. 1991;84:520–526. 5. Mangano DT, Multicenter Study of Perioperative Ischemia Research G. Aspirin and mortality from coronary bypass surgery. N Engl J Med. 2002;347:1309–1317. 6. Mehta RH, Sheng S, O’Brien SM, et al. Reoperation for bleeding in patients undergoing coronary artery bypass surgery: incidence, risk factors, time trends, and outcomes. Circ Cardiovasc Qual Outcomes. 2009;2:583–590. 7. Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006;34:1608–1616. 8. Ranucci M, Bozzetti G, Ditta A, et al. Surgical reexploration after cardiac operations: why a worse outcome? Ann Thorac Surg. 2008;86:1557–1562. 9. Sun J, Crowther M, Warkentin TE, et al. Should aspirin be discontinued before coronary artery bypass surgery? Circulation. 2005;112:e85–e90. 10. Hockings BEF, Ireland MA, Gotch-Martin KF, et al. Placebo-controlled trial of enteric coated aspirin in coronary bypass graft patients. Med J Aust. 1993;159:376–378. 11. Vuylsteke A, Oduro A, Cardan E, Latimer RD. Effect of aspirin in coronary artery bypass grafting. J Cardiothorac Vasc Anesth. 1997;11:831–834. 12. Sun J, Whitlock R, Cheng J, et al. 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;29:1057–1071. 13. Burger W, Chemnitius JM, Kneissl GD, et al. Low-dose aspirin for secondary cardiovascular prevention: cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation: review and meta-analysis. J Intern Med. 2005;257:399–414. 14. Bybee KA, Powell BD, Valeti U, et al. Preoperative aspirin therapy is associated with improved postoperative outcomes in patients undergoing coronary artery bypass grafting. Circulation. 2005;112:I286–I292. 15. Dacey LJ, Munoz JJ, Johnson ER, et al. Northern New England Cardiovascular Disease Study Group. Effect of preoperative aspirin use on mortality in coronary artery bypass grafting patients. Ann Thorac Surg. 2000;70:1986– 1990. 16. Ferraris VA, Ferraris SP, Moliterno DJ, et al. The Society of Thoracic Surgeons practice guideline series: aspirin and other antiplatelet agents during operative coronary revascularization (executive summary). Ann Thorac Surg. 2005;79:1454–1461. 17. Eagle KA, Guyton RA, Davidoff R, et al. 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). Circulation. 2004;110:1168–1176. 18. Theroux P, Waters D, Qiu S, et al. Aspirin versus heparin to prevent myocardial infarction during the acute phase of unstable angina. Circulation. 1993;88:2045–2048. 19. Verheugt FW, van der Laarse A, Funke-Kupper AJ, et al. Effects of early intervention with low-dose aspirin (100 mg) on infarct size, reinfarction and mortality in anterior wall acute myocardial infarction. Am J Cardiol. 1990;66:267–270. 20. Li S, Price R, Phiroz D, et al. Systemic inflammatory response during cardiopulmonary bypass and strategies. J Extra Corpor Technol. 2005;37: 180–188. 21. Lavi S, Lavi R. Perioperative management of antiplatelet agents in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2012;26:680–686. 22. Weerasinghe A, Taylor KM. The platelet in cardiopulmonary bypass. Ann Thorac Surg. 1998;66:2145–2152. 23. Zimmermann N, Gams E, Hohlfeld T. Aspirin in coronary artery bypass surgery: new aspects of and alternatives for an old antithrombotic agent. Eur J Cardiothorac Surg. 2008;34:93–108. 24. Jacob M, Smedira N, Blackstone E, et al. Effect of timing of chronic preoperative aspirin discontinuation on morbidity and mortality in coronary artery bypass surgery. Circulation. 2011;123:577–583. 25. Jimenez AH, Stubbs ME, Tofler GH, et al. Rapidity and duration of platelet suppression by enteric-coated aspirin in healthy young men. Am J Cardiol. 1992;69:258–262. 26. Krasopoulos G, Brister SJ, Beattie WS, et al. Aspirin ‘‘resistance’’ and risk of cardiovascular morbidity: systematic review and meta-analysis. BMJ. 2008;336:195–198.
www.annalsofsurgery.com | 1155
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Deng et al
27. Crescente M, Di Castelnuovo A, Iacoviello L, et al. Response variability to aspirin as assessed by the platelet function analyzer (PFA)-100. A systematic review. Thromb Haemost. 2008;99:14–26. 28. Kestin AS, Valeri CR, Khuri SF, et al. The platelet function defect of cardiopulmonary bypass. Blood. 1993;82:107–117. 29. Morse DS, Adams D, Magnani B. Platelet and neutrophil activation during cardiac surgical procedures: impact of cardiopulmonary bypass. Ann Thorac Surg. 1998;65:691–695. 30. Weightman WN, Gibbs NM, Weidmann CR, et al. The effect of preoperative aspirin-free interval on red blood cell transfusion requirements in cardiac surgical patients. J Cardiothorac Vasc Anesth. 2002;16:54–58.
1156 | www.annalsofsurgery.com
31. Munoz JJ, Birkmeyer NJ, Dacey LJ, et al. Trends in rates of reexploration for hemorrhage after coronary artery bypass surgery. Ann Thorac Surg. 1999;68:1321–1325. 32. Cannon CP, Mehta SR, Aranki SF. Balancing the benefit and risk of oral antiplatelet agents in coronary artery bypass surgery. Ann Thorac Surg. 2000;80:768–779. 33. Antiplatelet Trialists C. Collaborative overview of randomized trials of antiplatelet therapy-I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ. 1994;308:81–106. 34. Berger JS, Sallum RH, Katona B, et al. Is there an association between aspirin dosing and cardiac and bleeding events after treatment of acute coronary syndrome? A systematic review of the literature. Am Heart J. 2012;164:153–162.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Association Between Preoperative Aspirin-dosing Strategy and Mortality After Coronary Artery Bypass Graft Surgery Yi Deng, MD, Paul V. Pisklak, MD, Vei-Vei Lee, MS,y Daniel A. Tolpin, MD, z Charles D. Collard, MD, z MacArthur A. Elayda, MD, PhD,y Joseph Coselli, MD,§ and Wei Pan, MD z
Objective: To determine whether preoperative aspirin—acetylsalicylic acid (ASA)—timing or dose independently affects 30-day all-cause mortality. Background: Preoperative ASA administration is associated with reduced morbidity and mortality after coronary artery bypass graft (CABG). However, data are lacking regarding optimal timing and dosing of ASA. Methods: We retrospectively reviewed data from 3018 consecutive patients who underwent CABG surgery between July 2005 and May 2011. Patients were assigned to 3 groups according to the time of the last preoperative ASA dose: (1) 24 hours or less preoperatively (n ¼ 1173), (2) between 24 and 72 hours (n ¼ 876), and (3) more than 72 hours or none (n ¼ 969). In a separate analysis, patients were grouped according to ASA dose: 81 mg (n ¼ 1285), 325 mg (n ¼ 1004), and none (n ¼ 543). The primary outcome was 30-day allcause mortality. Results: The 30-day mortality rate was significantly lower in patients who took ASA 24 hours or less preoperatively (1.5%) than in those who took it between 24 and 72 hours (3.2%) or more than 72 hours or none (2.9%). Multivariate analysis showed that ASA within 24 hours preoperatively was associated with reduced mortality (odds ratio [OR], 0.41; 95% confidence interval [CI], 0.20–0.82; P ¼ 0.01). Moreover, mortality was significantly reduced for patients taking 81 mg of ASA (1.4%) compared with 325 mg (2.9%) or none (3.9%). Multivariate analysis demonstrated that 81 mg of ASA decreased mortality risk by 66% (OR, 0.34; 95% CI, 0.18–0.66; P < 0.01), whereas 325 mg of ASA had no mortality benefit (OR, 0.74; 95% CI, 0.41– 1.35; P ¼ 0.33) compared with no ASA. Conclusions: Low-dose ASA use within 24 hours of CABG is independently associated with decreased early postoperative mortality.
bleeding risk have examined the continued use of ASA up to the time of surgery; in most studies, ASA was stopped 5 to 7 days preoperatively.12 Prematurely discontinuing ASA can lead to a prothrombotic state in patients who are already at higher risk of ischemia to vital organs.13 Two retrospective studies demonstrated that preoperative ASA use, despite showing a trend toward increased bleeding and transfusion, improved surgical outcomes including mortality.14,15 Furthermore, a large prospective study demonstrated that stopping ASA before CABG was an independent risk factor for increased mortality.5 Because the bleeding risk can vary significantly depending on the time and dose of last taken, current guidelines cite a wide range (from 3 to 10 days) for the optimal time to discontinue ASA before cardiac surgery.16,17 Furthermore, most studies of preoperative ASA in CABG surgery were not powered to examine important perioperative outcomes such as mortality, and they varied widely in ASA dose. To our knowledge, there is no published comparative study examining the effect of the interval of ASA discontinuation or the dose taken before elective CABG on 30-day all-cause mortality. Thus, the aim of this study was to determine whether the time of discontinuation or the dose of preoperative ASA was independently associated with a reduced incidence of 30-day all-cause mortality after primary on-pump CABG. To this end, we performed a multivariate stepwise logistic regression analysis on data from more than 3000 consecutive patients undergoing primary CABG at our institution, controlling for patient demographics, medical history, and preoperative medication.
Keywords: aspirin, coronary artery bypass graft, mortality
METHODS
(Ann Surg 2015;262:1150–1156)
A
spirin—acetylsalicylic acid (ASA)—has become a mainstay treatment in both primary and secondary prevention of patients with cardiovascular disease.1,2 In coronary artery bypass grafting (CABG), early postoperative ASA increases graft patency, reduces ischemic complications, improves survival, and is recommended within hours of surgery.3 –5 However, preoperative ASA use remains controversial. Many surgeons recommend patients discontinue ASA 7 days before CABG for fear of excessive bleeding, which can lead to increased transfusion requirement, greater complications, and increased morbidity and mortality.6 –11 However, evidence in the literature is conflicting, and few studies of preoperative ASA and From the Department of Anesthesiology, Baylor College of Medicine, Houston, TX; yDepartment of Biostatistics and Epidemiology, Texas Heart Institute, Houston, TX; zDivision of Cardiovascular Anesthesiology, Texas Heart Institute, Baylor St. Luke’s Medical Center, Houston, TX; and §Department of Cardiovascular Surgery, Texas Heart Institute, Houston, TX. Disclosure: The authors declare no conflicts of interest. Reprints: Wei Pan, MD, Division of Cardiovascular Anesthesiology, Texas Heart Institute, 6720 Bertner Ave, Houston, TX 77030. E-mail: [email protected]. Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0003-4932/14/26105-0821 DOI: 10.1097/SLA.0000000000000951
1150 | www.annalsofsurgery.com
Study Design In this retrospective cohort study, we reviewed data from consecutive patients undergoing primary isolated CABG surgery with cardiopulmonary bypass (n ¼ 3018) between July 2005 and May 2011 at the Texas Heart Institute (THI). We grouped patients according to the interval between their last dose of ASA and the time of surgery: (1) those who took ASA 24 hours or less before surgery (n ¼ 1173), (2) those who took ASA between 24 and 72 hours before surgery (n ¼ 876), and (3) those who took ASA more than 72 hours before surgery or did not take it at all (n ¼ 969). In a separate analysis, patients were also analyzed on the basis of ASA dose: 81 mg (n ¼ 1285), 325 mg (n ¼ 1004), and none (n ¼ 543). Because of their limited number, patients taking 162 mg of ASA (n ¼ 35) were excluded from the analysis. We also excluded patients who underwent concomitant valve or other cardiac surgery (eg, septal defect repair and ventricular aneurysm resection). The study was approved by the institutional review board at Baylor St. Luke’s Medical Center.
Data Collection Patient demographics, preoperative risk factors and medications, and the incidence of adverse postoperative outcomes were obtained from the cardiac surgical database at the THI. The timing Annals of Surgery Volume 262, Number 6, December 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Preoperative ASA in CABG and Mortality
and dosage of ASA were gathered from patients’ chart when available. For those who were admitted to the hospital on the day of surgery, the dose and time of the last ASA taken were obtained from a preoperative medication questionnaire that was completed as part of the standard preadmission work-up for each patient. For those who were already inpatients, the nursing medication administration record was used to gather the time and dose of the last ASA before surgery. The measured adverse outcomes included 30-day all-cause mortality, myocardial infarction (MI), stroke, and ventilator dependency. A diagnosis of MI was made if there were new Q waves, new persistent ST-segment or T-wave changes, or evidence of acute MI on autopsy. A diagnosis of stroke was made if there was clinical evidence or evidence of a focal or global defect on computed tomography, magnetic resonance imaging, or autopsy. Ventilator dependency was defined as mechanical ventilation for more than 48 hours. Although we assessed the incidence of reoperation for bleeding, the quantitative blood loss during initial surgery or the units and types of transfused blood products were not available in the database.
associated with a reduction in the risk of 30-day all-cause mortality after primary CABG, we performed a multivariate, stepwise, logistic regression after adjusting for potentially confounding variables. Odds ratios and corresponding 95% confidence intervals (CIs) are reported with the associated P values. To further control for selection bias, we used 27 perioperative variables to create propensity-matched populations for 2 separate analyses. In the first analysis, patients receiving ASA 24 hours or less before surgery were compared with all others, and in the second analysis, patients receiving 81 mg of ASA were compared with all others. Propensity scores were calculated for each patient, and patients were matched one-to-one between the 2 groups. After successful matching, we performed a multivariate, stepwise logistic regression to determine whether ASA timing and dose were independently associated with a reduction in the risk of 30-day all-cause mortality after primary CABG surgery.
Statistical Analysis All statistical analyses were performed using SAS statistical software (v9.1, SAS Institute, Inc, Cary, NC) by the Division of Biostatistics and Epidemiology at the THI. Categorical (frequency) variables are expressed as percentages, and continuous variables are expressed as the mean standard deviation. We performed a univariate analysis using the x2 test for categorical data and a t test for continuous variables to compare preoperative patient characteristics and medication, perioperative risk factors, and perioperative outcomes, with the primary endpoint being 30-day all-cause mortality. To determine whether ASA timing or dose was independently
RESULTS Between July 2005 and May 2011, 3018 patients underwent elective primary CABG at the THI. Of the total group of patients, 39% took ASA 24 hours or less before surgery, 29% discontinued it between 24 to 72 hours, and 32% either stopped it more than 72 hours before surgery or never took it (Table 1). Overall, the 3 groups were similar with regard to age, the number of men, and the history of smoking, obesity, hypertension, peripheral vascular disease, previous stroke, renal insufficiency, and diabetes (Table 1). However, patients who had a previous MI, unstable angina at the time of admission, urgent surgery, or worse heart failure symptoms (New York Heart Association class III/IV) took ASA more frequently within 72 hours of surgery. Patients who continued ASA closer to surgery were less likely to be on an intra-aortic balloon pump. Concurrent b-blocker
TABLE 1. Perioperative Demographics and Risk Factors for Patients Grouped by ASA Timing Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass (%) Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Clopidogrel, % Previous IABP, % Cross-clamp time, min Total bypass time, min
24 hrs (n ¼ 1173)
24 to 72 hrs (n ¼ 876)
>72 hrs or None (n ¼ 969)
P
49.3 11.9 77.5 29.6 66.3 32.1 42.8 56.8 12.0 48.9 38.7 5.9 87.6 23.4 52.6 28.5 86.9 43.7 6.8 35.0 21.6 59.7 52.3 27.0 5.8 40.0 66.2
48.9 11.8 77.4 29.4 67.2 33.6 44.6 60.3 12.9 50.5 41.3 8.0 87.6 23.0 52.6 27.7 85.1 47.0 8.5 34.6 24.0 56.6 50.1 25.7 6.1 41.2 68.2
50.2 10.2 73.5 29.2 63.8 30.0 35.9 52.7 10.0 37.0 30.8 7.4 88.0 24.5 52.1 26.8 83.9 45.5 8.0 32.9 22.8 55.6 49.2 30.3 6.9 38.9 65.2
NS NS NS NS NS NS < 0.01 < 0.01 NS < 0.01 < 0.01 NS NS NS NS NS NS NS NS NS NS NS NS NS < 0.01 NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intra-aortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1151
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Deng et al
30-Day All-Cause Mortality (%)
15.0
10.0
5.0
n = 876 n = 1173*
n = 969
3.2
2.9
1.5 0.0
≤ 24 hours
> 24 hours - ≤ 72 hours > 72 hours + none ASA Taken in Hours Prior to Surgery
*P < 0.05 in multivariate analysis performed for 30-day mortality.
FIGURE 1. Incidence of 30-day mortality in patients undergoing CABG on the basis of the time of discontinuation of aspirin before surgery. P < 0.05 in multivariate analysis performed for 30-day mortality. and angiotensin-converting enzyme (ACE) inhibitor use did not differ among the groups. There were no significant differences in ASA use before 2006, from 2007 to 2008, and after 2009 (P ¼ 0.20; data not shown). When stratified by the time of discontinuation of ASA before surgery, the incidence of 30-day all-cause mortality was 1.5% in patients who stopped ASA 24 hours or less before surgery, 3.2% in those who stopped between 24 and 72 hours, and 2.9% in those who stopped more than 72 hours before surgery or never took ASA (Fig. 1). In a multivariate analysis controlling for patient demographics, preoperative risk factors, and intraoperative variables, we found that taking ASA within 24 hours of surgery was independently associated with reduced mortality [odds ratio (OR), 0.41; 95% CI, 0.20–0.82; P ¼ 0.01]. There was no significant difference between the groups in other outcomes such as incidence of MI, stroke, ventilator dependency, or reoperation for bleeding (Table 2). In addition, the 30-day all-cause mortality of surgery was 2.5% in 2005 to 2006, 2.4% in 2007 to 2008, and 2.4% in 2009 to 2011. There were no significant differences in the death rate among these years (P ¼ 0.98). To assess the effect of preoperative ASA dose on mortality, we separated the patients into 3 groups: low dose (81 mg), high dose (325 mg), or no ASA (Table 3). Overall, 45% of patients were on low dose, 35% on high dose, and 19% did not take ASA. Predictably, patients who had a previous MI, urgent surgery, unstable angina, and worse heart failure symptoms were more likely to be taking ASA preoperatively, especially the dose of 325 mg. Moreover, the incidence of hyperlipidemia was greater in patients receiving preoperative ASA than in those receiving no ASA. However, the incidence of hypertension, stroke, diabetes, or smoking was similar between groups. Both groups of ASA takers were also more likely to be on b-blockers and ACE inhibitors at the same time. Interestingly,
older patients (age >65 years) were less likely to be on high-dose ASA than on low-dose or no ASA. Figure 2 shows the incidence of 30-day all-cause mortality according to ASA dose. Mortality was 1.4% for patients taking 81 mg of ASA, 2.9% for those taking 325 mg of ASA, and 3.9% for those not taking ASA (P < 0.01). Multivariate analysis showed that mortality was 66% lower in patients on low-dose ASA than in those not receiving preoperative ASA (OR, 0.34; 95% CI, 0.18–0.66; P < 0.01), whereas no mortality benefit was seen with high-dose ASA (OR, 0.74; 95% CI, 0.41–1.35; P ¼ 0.33). The incidence of postoperative MI and stroke did not differ among all 3 groups (Table 4). Preoperative ASA did not result in higher rates of reoperation for bleeding compared with no ASA, and the rates of reoperation were similar for the 325-mg and 81mg groups (3.8% vs 4.1%, respectively). To further control for selection bias related to the choice of therapy, we used 27 variables to create propensity-matched populations for 2 separate analyses. In the first cohort, we successfully matched patients on a one-to-one basis who discontinued ASA use within 24 hours of surgery (n ¼ 1173) with those who discontinued it more than 24 hours after surgery (n ¼ 1173) (Table 5). In the second cohort, patients who took 81 mg of ASA preoperatively were matched with all other patients who were not taking low-dose ASA (n ¼ 1053 for each group) (Table 6). We then performed stepwise logistic regression on the matched cohorts. Even after controlling for propensity scores, the observed reduction in 30day all-cause mortality associated with low-dose ASA use persisted in the multivariate analysis. Other significant predictors of early mortality included older age (OR, 2.3; 95% CI, 1.2–4.5; P ¼ 0.03), previous MI (OR, 2.7; 95% CI, 1.4–5.0; P < 0.01), and preoperative renal dysfunction (OR, 4.2; 95% CI, 2.2–7.7; P < 0.01) (Fig. 3).
DISCUSSION In this single-center study of preoperative ASA use in 3018 patients undergoing primary isolated CABG, we found that taking ASA within 24 hours of surgery was associated with a 48% relative reduction in 30-day all-cause mortality. This finding held true even when we adjusted for confounding risk factors, severity of illness, and other medication use. In a separate analysis, we found that preoperative low-dose (81 mg) ASA was independently associated with a reduced 30-day mortality rate when compared with no ASA, but preoperative high-dose (325 mg) ASA did not confer a mortality benefit. These data suggest that preoperative low-dose ASA therapy within 24 hours of surgery would have a favorable effect on patients undergoing primary CABG surgery. ASA therapy has been shown to reduce mortality in the setting of acute coronary syndrome.18,19 Although the mechanism underlying the association of preoperative ASA and reduced surgical mortality is unknown, increased platelet activation during CABG surgery can lead to thrombotic or ischemic events. Thus, it is reasonable to postulate that ASA can confer a mortality benefit by limiting platelet activation, thereby reducing graft thrombosis as well as other cerebrovascular, gastrointestinal, or renal
TABLE 2. Incidence of Adverse Postoperative Outcomes as Grouped by ASA Timing 24 hrs (n ¼ 1173)
24 to 72 hrs (n ¼ 876)
>72 hrs or None (n ¼ 969)
P
1.5 3.1 3.9 2.5
3.2 4.3 5.4 2.6
2.9 3.9 3.4 2.6
0.02 NS NS NS
30-d mortality, % MI, % Reoperation for bleeding, % Stroke, % NS indicates not significant.
1152 | www.annalsofsurgery.com
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Preoperative ASA in CABG and Mortality
TABLE 3. Perioperative Demographics and Risk Factors for Patients Grouped by ASA Dose Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Clopidogrel, % Previous IABP, % Cross-clamp time, min Total bypass time, min
None (n ¼ 543)
81 mg (n ¼ 1285)
325 mg (n ¼ 1004)
P
50.1 10.7 72.4 29.6 65.7 31.8 39.0 42.1 10.1 42.1 29.1 8.1 85.4 22.1 50.3 28.1 81.0 45.0 7.4 34.1 21.7 52.5 45.4 21.5 5.7 39.6 67.4
52.6 11.2 77.5 29.2 63.9 30.2 37.9 42.7 11.5 42.7 34.0 7.6 88.8 24.1 52.2 25.9 87.9 44.6 6.8 31.6 23.3 60.6 53.6 32.8 4.4 39.8 65.6
45.0 12.0 77.3 29.7 68.3 33.9 46.8 51.7 13.0 51.7 46.4 5.5 88.0 24.0 54.3 29.7 85.3 46.4 9.1 36.3 22.5 57.1 50.9 25.7 6.2 40.6 66.8
< 0.01 NS 0.02 NS NS NS < 0.01 0.03 NS < 0.01 < 0.01 NS NS NS NS NS < 0.01 NS NS NS NS < 0.01 < 0.01 < 0.01 NS NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intra-aortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
complications.5,20–23 Our study showed a decrease in early mortality but not a reduction in postoperative MI or stroke. These results are similar to those of other outcome studies of preoperative ASA in CABG surgery, suggesting that ASA can reduce post-CABG mortality in ways other than decreasing cardiac or cerebrovascular complications.14,24 Because our institution did not routinely examine postoperative cardiac enzyme levels, our ability to detect postoperative MI or cardiac dysfunction was limited. Thus, we focused on the well-defined clinical endpoint of early all-cause mortality as our primary outcome. In our study, preoperative low-dose ASA continued up to the time of surgery significantly decreased short-term postsurgical
30-Day All-Cause Mortality (%)
15.0
10.0
5.0
n = 543 n = 1004
3.9
n = 1285*
2.9
TABLE 4. Incidence of Adverse Postoperative Outcomes as Grouped by ASA Dose
1.4
0.0
None
81 mg ASA Dose Prior to Surgery
325 mg
*P < 0.05 in multivariate analysis performed for 30-day mortality.
FIGURE 2. Incidence of 30-day mortality in patients undergoing CABG on the basis of aspirin dose before surgery. P < 0.05 in multivariate analysis performed for 30-day mortality. ß
mortality. To our knowledge, this is the only study that specifically compared intervals of ASA use as close as 24 hours within surgery. In most comparative studies, ASA use is discontinued 7 days before surgery, which likely underestimates its potential benefits. The clinical effect of ASA is typically shorter than 7 days, and some studies suggest that platelet inhibition is minimal after 3 days and can vary significantly among individuals.25–27 Also, surgeons have been reluctant to continue ASA closer than 7 days before surgery for fear of increased bleeding, given that cardiopulmonary bypass can further decrease platelet function postoperatively through hypothermia, sequestration, and mechanical factors.28,29 This concern was supported by studies published in the 1980 s to early 1990 s that showed significantly increased bleeding and transfusion-related complications associated with preoperative ASA use in CABG surgery.12 However, over the last 2 decades, changes in clinical practices may have addressed some of these issues, tipping the scales in favor of preoperative ASA use. First, modern practices of aggressive bloodsalvaging techniques and the widespread adoption of antifibrinolytics such as e-aminocaproic acid may have blunted any additional
2015 Wolters Kluwer Health, Inc. All rights reserved.
30-d mortality, % MI, % Reoperation for bleeding, % Stroke, %
None (n ¼ 543)
81 mg (n ¼ 1285)
325 mg (n ¼ 1004)
P
3.9 4.8 4.8 2.7
1.4 3.1 4.1 2.4
2.9 3.7 3.8 2.6
< 0.01 NS NS NS
NS indicates not significant.
www.annalsofsurgery.com | 1153
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Deng et al
Annals of Surgery Volume 262, Number 6, December 2015
TABLE 5. Demographics of Propensity-matched Groups by the Time of ASA Discontinuation Before Surgery
TABLE 6. Demographics of Propensity-matched Groups by ASA Dose Before Surgery
Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % Pulmonary disease, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Statin use, % Previous IABP, % Cross-clamp time, min Total bypass time, min
24 hrs (n ¼ 1173)
>24 hrs (n ¼ 1173)
P
49.3 11.9 67.5 29.6 66.3 32.1 42.8 56.8 12.0 48.9 38.7 5.9 87.6 34.7 23.4 52.6 28.5 86.9 43.7 6.8 35.0 21.6 59.7 52.3 81.7 4.2 40.0 66.2
49.7 11.5 65.2 29.2 64.5 31.5 39.8 56.7 11.5 48.9 37.4 7.5 88.5 38.9 23.3 52.6 27.0 85.7 46.1 9.0 33.6 22.4 58.9 50.8 81.7 5.8 40.5 67.0
NS NS NS NS NS NS NS NS NS NS NS NS NS 0.04 NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Age >65 yrs, % African Americans, % Male, % Body mass index, kg/m2 Three-vessel bypass, % Left main disease, % Urgent surgery, % NYHA class III/IV, % Low ejection fraction, % Unstable angina, % Previous MI, % Valvular disease, % Hypertension, % Pulmonary disease, % PVD, % Smoking, % Obesity, % Hyperlipidemia, % Diabetes, % Previous CVA, % Family history of CAD, % Renal insufficiency, % b-Blocker use, % ACE inhibitors, % Statin use, % Previous IABP, % Cross-clamp time, min Total bypass time, min
81 mg (n ¼ 1053)
Not on 81 mg (n ¼ 1053)
P
48.6 11.9 78.2 29.3 67.5 32.1 40.2 57.0 12.1 45.5 36.6 7.1 88.7 36.6 24.0 52.3 27.0 87.4 44.3 4.8 34.2 22.6 59.2 53.9 81.1 5.2 40.7 67.5
48.6 11.2 74.8 29.6 68.1 31.2 39.7 57.3 11.7 45.0 37.1 6.4 88.7 35.3 23.1 53.0 29.2 85.7 47.7 5.4 35.0 23.7 58.8 51.6 81.7 5.1 40.2 66.6
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intraaortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
CAD indicates coronary artery disease; CVA, cerebrovascular accident; IABP, intraaortic balloon pump; NS, not significant; NYHA, New York Heart Association; PVD, peripheral vascular disease.
bleeding effect from ASA. Moreover, recent studies have shown that stopping ASA within 3 days did not result in increased bleeding when compared with 7 days.11,30 This may enable patients to continue receiving the benefits of ASA therapy until surgery without the concern for increased bleeding complications.31,32 Second, when compared with the earlier studies, patients today are typically older, present with more comorbidities, are more likely to have coronary
stents, and are on chronic ASA therapy.5,12,31 ASA withdrawal before surgery can be associated with a 2% to 10% incidence of acute coronary syndrome and a 6% incidence of acute lower-limb ischemia.13 Thus, in today’s patients, the risk of ASA withdrawal may outweigh the risk of bleeding complications and may even adversely affect postoperative mortality. In fact, 2 large retrospective studies from this decade showed discontinuing ASA 5 to 7 days before
Independent Predictors
95% CI
PVD
1.2-4.0
Renal dysfunction
2.2-7.7
Previous MI
1.4-5.0
NYHA class III/IV
1.1-4.8
Female
1.1-3.0
Age > 65
1.2-4.5
ASA ≤ 24 hours
0.2-0.8
ASA 81 mg
0.2-0.7 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Odds Ratio (95% CI)
1154 | www.annalsofsurgery.com
ß
FIGURE 3. Multivariate analysis of independent predictors on 30-day all-cause mortality. PVD indicates peripheral vascular disease.
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
CABG was associated with increased postoperative mortality.14,15 Our study suggests that stopping ASA more than 24 hours before surgery may lead to worse outcomes. The data on the optimal dose of ASA before CABG surgery are limited because there are no trials that directly compare different dosing regimens and their outcomes. In preventing vein graft occlusions after CABG, low-dose ASA (75–325 mg) was not inferior to high-dose ASA (500–1500 mg).33 In interventional cardiology procedures for acute coronary syndrome, data demonstrate that higherdose ASA did not decrease the incidence of MI and all-cause mortality but did significantly increase bleeding complications.34 In a metaanalysis by Sun and colleagues12 examining the effect of preoperative ASA in CABG, the dose of ASA included in studies varied significantly between 80 and 2600 mg. They reported that the use of doses of 325 mg or more was associated with a significant increase in bleeding, whereas doses of less than 325 mg were not. The meta-analysis was not powered to examine the dose effect on mortality. Thus, it seems that given the equivocal benefit and the clearly identified risks of using high-dose ASA, low-dose ASA may be more beneficial for patients undergoing CABG. Our study demonstrated that low-dose (81 mg) ASA was independently associated with surgical mortality, whereas high-dose (325 mg) was not. It may be that the increased bleeding complications from high-dose ASA negated its benefits. However, we were unable to assess that hypothesis because of the lack of quantitative blood loss and transfusion-related data in our database. Our study has several limitations. First, this is a retrospective study, so we could not control for the administration of ASA preoperatively. Despite careful use of regression models to adjust for potential confounders, many factors may still affect the outcome. However, we attempted to limit bias by using propensity matching with a large number of variables to ensure the groups were as evenly matched as possible with regard to demographics and perioperative risk factors. Second, this study was conducted at a single institution; however, our perioperative outcomes were similar to recently published results from other large centers for cardiac surgery, suggesting that the experiences were comparable.14,15 Third, our database did not track the amount of bleeding or transfusion requirements, so we were unable to examine the relationship between preoperative ASA and increased bleeding complications. Nevertheless, the rates of reoperations for bleeding were similar in all groups and were comparable to recently published data, suggesting that the bleeding and transfusion requirements may not have differed significantly.14,15,30,31 Lastly, we did not measure or control for the effect of postoperative ASA administration, although standard practice at the THI is for all patients to begin ASA within 1 day after surgery.
CONCLUSIONS The timing and dose of ASA before CABG surgery are associated with the 30-day mortality rates. Specifically, ASA use (81 mg) within 24 hours of surgery is independently associated with decreased postoperative mortality. Future large-scale, randomized, controlled studies are warranted to further identify the role of preoperative ASA in perioperative outcome benefits in cardiac surgical patients.
ACKNOWLEDGMENTS We thank Rebecca Bartow, PhD, of the Texas Heart Institute, for editorial assistance.
REFERENCES 1. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk, patients. BMJ. 2002;324:71–86. 2. The´roux P, Ouimet H, McCans J, et al. Aspirin, heparin, or both to treat acute unstable angina. N Engl J Med. 1988;319:1105–1111.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Preoperative ASA in CABG and Mortality
3. Chesebro JH, Clements IP, Fuster V, et al. A platelet-inhibitor drug trial in coronary-artery bypass operations. N Engl J Med. 1982;307:73–78. 4. Goldman S, Copeland J, Moritz T, et al. Starting aspirin therapy after operation: effects on early graft patency. Department of Veterans Affairs Cooperative Study Group. Circulation. 1991;84:520–526. 5. Mangano DT, Multicenter Study of Perioperative Ischemia Research G. Aspirin and mortality from coronary bypass surgery. N Engl J Med. 2002;347:1309–1317. 6. Mehta RH, Sheng S, O’Brien SM, et al. Reoperation for bleeding in patients undergoing coronary artery bypass surgery: incidence, risk factors, time trends, and outcomes. Circ Cardiovasc Qual Outcomes. 2009;2:583–590. 7. Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006;34:1608–1616. 8. Ranucci M, Bozzetti G, Ditta A, et al. Surgical reexploration after cardiac operations: why a worse outcome? Ann Thorac Surg. 2008;86:1557–1562. 9. Sun J, Crowther M, Warkentin TE, et al. Should aspirin be discontinued before coronary artery bypass surgery? Circulation. 2005;112:e85–e90. 10. Hockings BEF, Ireland MA, Gotch-Martin KF, et al. Placebo-controlled trial of enteric coated aspirin in coronary bypass graft patients. Med J Aust. 1993;159:376–378. 11. Vuylsteke A, Oduro A, Cardan E, Latimer RD. Effect of aspirin in coronary artery bypass grafting. J Cardiothorac Vasc Anesth. 1997;11:831–834. 12. Sun J, Whitlock R, Cheng J, et al. 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;29:1057–1071. 13. Burger W, Chemnitius JM, Kneissl GD, et al. Low-dose aspirin for secondary cardiovascular prevention: cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation: review and meta-analysis. J Intern Med. 2005;257:399–414. 14. Bybee KA, Powell BD, Valeti U, et al. Preoperative aspirin therapy is associated with improved postoperative outcomes in patients undergoing coronary artery bypass grafting. Circulation. 2005;112:I286–I292. 15. Dacey LJ, Munoz JJ, Johnson ER, et al. Northern New England Cardiovascular Disease Study Group. Effect of preoperative aspirin use on mortality in coronary artery bypass grafting patients. Ann Thorac Surg. 2000;70:1986– 1990. 16. Ferraris VA, Ferraris SP, Moliterno DJ, et al. The Society of Thoracic Surgeons practice guideline series: aspirin and other antiplatelet agents during operative coronary revascularization (executive summary). Ann Thorac Surg. 2005;79:1454–1461. 17. Eagle KA, Guyton RA, Davidoff R, et al. 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). Circulation. 2004;110:1168–1176. 18. Theroux P, Waters D, Qiu S, et al. Aspirin versus heparin to prevent myocardial infarction during the acute phase of unstable angina. Circulation. 1993;88:2045–2048. 19. Verheugt FW, van der Laarse A, Funke-Kupper AJ, et al. Effects of early intervention with low-dose aspirin (100 mg) on infarct size, reinfarction and mortality in anterior wall acute myocardial infarction. Am J Cardiol. 1990;66:267–270. 20. Li S, Price R, Phiroz D, et al. Systemic inflammatory response during cardiopulmonary bypass and strategies. J Extra Corpor Technol. 2005;37: 180–188. 21. Lavi S, Lavi R. Perioperative management of antiplatelet agents in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2012;26:680–686. 22. Weerasinghe A, Taylor KM. The platelet in cardiopulmonary bypass. Ann Thorac Surg. 1998;66:2145–2152. 23. Zimmermann N, Gams E, Hohlfeld T. Aspirin in coronary artery bypass surgery: new aspects of and alternatives for an old antithrombotic agent. Eur J Cardiothorac Surg. 2008;34:93–108. 24. Jacob M, Smedira N, Blackstone E, et al. Effect of timing of chronic preoperative aspirin discontinuation on morbidity and mortality in coronary artery bypass surgery. Circulation. 2011;123:577–583. 25. Jimenez AH, Stubbs ME, Tofler GH, et al. Rapidity and duration of platelet suppression by enteric-coated aspirin in healthy young men. Am J Cardiol. 1992;69:258–262. 26. Krasopoulos G, Brister SJ, Beattie WS, et al. Aspirin ‘‘resistance’’ and risk of cardiovascular morbidity: systematic review and meta-analysis. BMJ. 2008;336:195–198.
www.annalsofsurgery.com | 1155
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 262, Number 6, December 2015
Deng et al
27. Crescente M, Di Castelnuovo A, Iacoviello L, et al. Response variability to aspirin as assessed by the platelet function analyzer (PFA)-100. A systematic review. Thromb Haemost. 2008;99:14–26. 28. Kestin AS, Valeri CR, Khuri SF, et al. The platelet function defect of cardiopulmonary bypass. Blood. 1993;82:107–117. 29. Morse DS, Adams D, Magnani B. Platelet and neutrophil activation during cardiac surgical procedures: impact of cardiopulmonary bypass. Ann Thorac Surg. 1998;65:691–695. 30. Weightman WN, Gibbs NM, Weidmann CR, et al. The effect of preoperative aspirin-free interval on red blood cell transfusion requirements in cardiac surgical patients. J Cardiothorac Vasc Anesth. 2002;16:54–58.
1156 | www.annalsofsurgery.com
31. Munoz JJ, Birkmeyer NJ, Dacey LJ, et al. Trends in rates of reexploration for hemorrhage after coronary artery bypass surgery. Ann Thorac Surg. 1999;68:1321–1325. 32. Cannon CP, Mehta SR, Aranki SF. Balancing the benefit and risk of oral antiplatelet agents in coronary artery bypass surgery. Ann Thorac Surg. 2000;80:768–779. 33. Antiplatelet Trialists C. Collaborative overview of randomized trials of antiplatelet therapy-I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ. 1994;308:81–106. 34. Berger JS, Sallum RH, Katona B, et al. Is there an association between aspirin dosing and cardiac and bleeding events after treatment of acute coronary syndrome? A systematic review of the literature. Am Heart J. 2012;164:153–162.
ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
