• 
• 

Three Part Question

In [patients following coronary arterial bypass graft] what [is the optimal dose of aspirin required] to [prolonging event free survival]?

Clinical Scenario

You are ready to discharge a 57-year-old gentleman who has undergone CABG 8 days ago. It is your consultant's policy to discharge all people without contraindications on low dose aspirin, but you have recently attended a structured critical appraisal journal club and wonder whether a higher dose of aspirin may confer a survival advantage to your patient.

Search Strategy

Medline 1966-08/03 using the OVID interface. Cochrane Central Register of Controlled Trials (CENTRAL).
([exp Coronary Artery Bypass OR coronary artery bypass.mp OR vascular graft.mp] AND [exp aspirin OR aspirin.mp] AND [exp graft Occlusion, Vascular OR graft occlusion.mp OR vascular patency.mp] AND [maximally sensitive randomised control trial filter])

Search Outcome

114 abstracts were found from Medline and 59 papers from CENTRAL. Two meta-analyses were found from 1993/4, which included all relevant studies pre-1991. Of the remaining papers, 104 were either irrelevant or of insufficient quality for inclusion and 4 further papers were relevant. The 4 papers and the 2 meta-analyses are shown in the table. The Journal Club also suggested one additional recent paper that had not yet been indexed by Medline, which is also included in the table.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Goldman et al, 1989, USA	N=406 patients receiving CABG with saphenous vein grafts Aspirin 325mg od (N=104) vs aspirin 325mg tds (N=96) vs aspirin and dipyridamole (325mg and 75mg o.d.) (N=99) vs placebo (N=107) Aspirin given 12 hours before operation and 6 hours after operation	Double blind PRCT	Graft patency at 9 days and 1 year after CABG as determined by angiography	325mg o.d. aspirin 13.2% occlusion 325mg t.d.s aspirin 16.8% occlusion Aspirin + dipyridamole 17.5% occlusion Placebo 22.6% occlusion	772 patients initially randomised, 502 had the 1 year angiogram (154 refusals, 32 lost to F/U, 31 deaths, 50 others) No difference in treatment subgroups and no difference in vessel larger than 2.0mm
				Occlusion rate in all aspirin groups 15.8% Occlusion rate in all placebo group 22.6% P=0.029
			Occlusion rate at 1 year in grafts patent at 9 days	If a graft was patent at 9 days the occlusion rate at 1 year was: All aspirin groups 10.0% Placebo groups 9.8% P=0.96
Goldman et al, 1990, USA	N= 237 IMA grafts to LAD and 383 vein grafts to LAD Patients after CABG with IMA and saphenous grafts Aspirin all regimes (see next paper) Vs placebo	Double blind PRCT	Graft patency at 9 days and 1 year on angiography of IMA to LAD	Aspirin groups 92.1% placebo group 100% P=0.385	Underpowered study as it is a sub-analysis of veterans study None of 23 IMAs who had placebo and only 17 of 214 IMAs who had aspirin occluded
			Graft patency at 9 days and 1 year on angiography of SV to LAD	Patency rate at 1 year of aspirin groups 90%% Patency rate of placebo group 88.8%% P=0.675
Fremes et al, 1993, Canada	12 studies that evaluated, occlusion rates of saphenous vein grafts after CABG. Aspirin in various regimes of 50mg to 975mg +/- Dipyridamole Vs Control N=3224 patients in the 12 trials	Meta-analysis	Proportion of patients with 1 or more distal anastomotic occlusions	Low dose <150mg OR 0.56 CI 0.45-0.69 Medium Dose 150-350mg OR 0.37 CI 0.25-0.53 High dose >350mg OR 0.70 CI 0.57-0.88	Saphenous veins only Significant Heterogeneity demonstrated amongst studies Unable to perform cluster analysis to assess significanceStudies are from July 1991 or earlier
			All cause mortality	No mortality benefit demonstrated with any regime
Antiplatelet Trialists' Collaboration, 1994, UK	46 studies evaluating presence of at lest one occlusion after a vascular procedure. 23 studies were CABG studies 13 studies provided data on aspirin medium dose (75-325mg) Vs High Dose (500-1500mg) N=3097	Meta-analysis	Prevention of at least one vascular occlusion assessed by angiography	Medium dose aspirin vs control Odds reduction 44% (SD 10%) High dose aspirin vs control Odds reduction 50% (SD 9%)	Not all studies included in meta-analysis are CABG studies, but it is unclear as to how many of these 13 studies are in CABG patients. Studies included are from March 1990 or earlier
			MI, stroke or vascular death	124/2529 patients in antiplatelet group, 127/2546 in adjusted controls No mortality benefit in CABG patients
Goldman et al, 1997, USA	N=266 male patients post CABG with saphenous veins, who had graft patency confirmed at 10 days by angiography, and 1 year of aspirin at 325mg Randomised to receive either 325mg (N=128) or placebo (N=130) for a further 2 years	Double blind PRCT	Graft patency at 3 years on angiography	Aspirin group 12.5% occlusion No aspirin group 15.3% occlusion P=0.323 Aspirin treatment from 1-3 years was not predictive of graft patency 3 years after CABG	Small number of occlusions, 16 vs 20 patients with occluded grafts 334 originally agreed to participate but they only have data on 266 at 3 years
Stein et al, 2001,	All studies considered after well conducted systematic review of the literature Of note this was the report of the 6th American College of Chest Physicians Task Force on Antithrombotic Therapy. They have been reviewing all evidence relating to antithrombotic and anti-coagulation therapies for 14 years	Systematic Review	Immediate postoperative regime after SV bypass operation	Recommend 325mg/d of aspirin, started 6hrs after operation. Grade 1A recommendation	Other meta-analyses were not considered The recommendation for lifelong aspirin in post CABG patients is based on mortality studies on patients with non-operated coronary arterial disease, which is a clearly different cohort to post-CABG patients
			Continuation of aspirin	Long-term aspirin does not increase vein patency, which is more related to operative technique, but is indicated as it reduces mortality in all patients with coronary disease. Grade 1A recommendation
			Low dose vs medium dose aspirin	There are no studies that compare 50-100mg/day to 325mg/day but there is also no evidence to support the concept of reduced GI bleeding with low dose aspirin
			Internal mammary arterial bypass grafting	Aspirin does not increase IMA patency. Aspirin should be given lifelong to all patients with a diagnosis of coronary arterial disease. Grade 1A
Mangano DT, 2002,	70 centres in 17 countries, prospectively studied. 5065 patients undergoing CABG. Pts receiving 80-650mg of aspirin 48hrs post-op. (N=2999) compared to those who did not(N=2023)	Multicentre prospective cohort study	All cause mortality	Aspirin group 40/2999 deaths (1.3%) Control group 81/2023 deaths (4.0%) P<0.001	50% of those who did not receive aspirin post-op were on aspirin preoperatively This paper did not sub-analyse patients receiving low-dose aspirin and medium dose aspirin
			Myocardial infarction	Aspirin group 80/2999 deaths (2.8%) Control group 105/2023 deaths (5.4%) P<0.001
			GI bleeding	Aspirin group 34/2999 deaths (1.1%) Control group 41/2023 deaths (2.0%) P=0.01

Comment(s)

Fremes' meta-analysis demonstrated a significant benefit of low and medium dose aspirin in comparison to high dose aspirin. The benefit of medium dose aspirin was greatest but confidence intervals cross the confidence intervals for low dose aspirin. Neither the anti-platelet trialists nor the veterans study group were able to convincingly demonstrate an advantage of medium dose aspirin in comparison to low dose aspirin. Mangano et al provide the first evidence for a convining mortality benefit from aspirin. However the range of Aspirin used was from 80mg to 650mg, so no evidence was provided for choosing a dose within this range. Of note they also found a higher rate of GI and bleeding complications in the non-aspirin group. These findings have all been supported By the ACCP Consensus Conference on Antithrombotic Therapy. They recommend 325mg/d as the current best evidence recommendation for aspirin post cardiac surgery.

Clinical Bottom Line

Best Evidence suggests that 325mg optimally improves vein graft survival and mortality, and does not cause an increase in complications compared to lower doses.

References

1. Goldman S, Copeland J, Moritz T, et al. Saphenous vein graft patency 1 year after coronary artery bypass surgery and effects of antiplatelet therapy. Results of a Veterans Administration Cooperative Study. Circulation 1989;80:1190-7.
2. Goldman S, Copeland J, Moritz T, et al. Internal mammary artery and saphenous vein graft patency. Effects of aspirin. Circulation 1990;82:IV237-IV242.
3. Fremes SE, Levinton C, Naylor CD, et al. Optimal antithrombotic therapy following aortocoronary bypass: a meta-analysis. European Journal of Cardio-Thoracic Surgery 1993;7:169-80.
4. Antiplatelet Trialists' Collaboration. Collaborative overview of randomised trials of antiplatelet therapy--II: Maintenance of vascular graft or arterial patency by antiplatelet therapy. BMJ 1994;308:159-68.
5. Goldman S, Zadina K, Krasnicka B, et al. Predictors of graft patency 3 years after coronary artery bypass graft surgery. Department of Veterans Affairs Cooperative Study Group No. 297. Journal of the American College of Cardiology 1997;29:1563-8.
6. Stein PD, Dalen JE, Goldman S, et al. Antithrombotic Therapy in Patients With Saphenous Vein and Internal Mammary Artery Bypass Grafts Chest 2001;119:278S-282S.
7. Mangano DT. Aspirin and mortality from coronary bypass surgery. NEJM 2002;347:1309-17.