Critically Appraised Topic:
Can low dose anticoagulation with warfarin and/or aspirin be effective in the primary prevention of ischemic heart disease in men at high risk?
Appraised by: Deb Bynum, MD
Date: Feb 11, 1998
Clinical Bottom Lines:
1) Low dose anticoagulation with warfarin (with a mean INR of 1.47) reduced all ischemic heart disease events (fatal and nonfatal) from 12.4% in patients not on warfarin (on aspirin alone or placebo) to 9.8% in patients on warfarin for a RRR of 21% (p=.02). This effect was primarily due to a 39% RRR in fatal events (4.8% of patients not on warfarin had a fatal IHD event compared to 2.9% patients on warfarin). This also accounted for an overall decrease in all cause mortality from 13.9% in patients not on warfarin to 11.6% -- a RRR of 17%. The effect of warfarin on nonfatal IHD events was not significant.
2) The use of aspirin (with or without warfarin) was associated with a 20% decrease in all IHD events (p=.04) from 11.8% in patients not on aspirin to 9.5% in those on aspirin. However, in contrast to warfarin, there was no difference in the rate of fatal IHD events (3.7% on aspirin vs 3.3% not on Aspirin). Nonfatal IHD events however were decreased from 8.5% to 5.8% with the use of aspirin (RRR 32%). There was no difference in all cause mortality between patients taking aspirin and those not on aspirin.
3) Warfarin and aspirin increased the risk of hemorrhagic and fatal strokes. Patients on warfarin had a slight increase in the rate of strokes from any cause (2.7% to 3.1%) and a small increase in hemorrhagic strokes from 0.2% to 0.5%. Aspirin was associated with a slight increase in hemorrhagic strokes from 0.1% to 0.6%, but a decrease in thrombotic strokes from 2.0% to 1.3% -- therefore there was no difference in the rate of strokes from any cause with aspirin (2.9% vs 3.0%).
4) There was an increased risk of ruptured aortic or dissecting aneurysm in patients on warfarin (15 patients) vs those not on warfarin (3 patients), p=.01.
5) There was a small but significant increased risk of major and minor bleeding episodes in patients on warfarin, however the risk was not significantly different than the rates seen with aspirin alone.
The Evidence: Randomised, blinded trial comparing patients on low dose warfarin, low dose aspirin, warfarin plus aspirin, or placebo alone with a primary endpoint of ischemic heart disease events (deaths from coronary causes or MI); Stroke and overall mortality were secondary endpoints. The patient population consisted of men at increased risk for heart disease, but no prior history of MI or strokes.
|warfarin + aspirin||warfarin||aspirin||placebo|
|All||71 (8.7%)||83 (10.3%)||83 (10.2%)||107 (13.3%)|
|Fatal||24 (3.0%)||19 (2.4%)||36 (4.4%)||34 (4.2%)|
|Nonfatal||47 (5.8%)||64 (8.0%)||47 (5.8 %)||73 (9.0%)|
|All cause||29 (3.6%)||22 (2.7%)||18 (2.2%)||26 (3.2%)|
|Thrombotic||11 (1.4%)||15 (1.9%)||10 (1.2%)||18 (2.2%)|
|Hemorrhagic||7 (0.9%)||1 (0.1%)||2 (0.2%)||0|
|Fatal||12 (1.5%)||5(0.6%)||2 (0.2%)||1 (0.1%)|
|Total Mortality||103 (12.4%)||95 (11.4%)||113 (13.6%)||110 (13.1%)|
1) Potential limitations to applying this to clinical practice include difficulty and cost of following patients on warfarin and issues of compliance. The authors point out that the process may be easier and safer with having the goal INR of 1.5.
2) Caution needed in patients with poorly controlled hypertension who seem to be at increased risk for strokes. The increased risk of aneurysms on warfarin raises concerns for need for screening prior to starting treatment which could be costly.
3) BIG POINT: need to weigh costs and benefits. The results are mainly given as relative risk reduction-- overall 5 IHD events could be avoided by treating 1000 men with warfarin and aspirin for one year ( or 3 events with warfarin alone, 3 with aspirin alone). In other words, the NNT with warfarin and aspirin to prevent on IHD event is 21. But, the all cause mortality in the placebo group was 13.1% compared to 12.4% in the warfarin plus aspirin group for an ARR of 0.7% -- The NNT for all cause mortality is therefore 143 ! -- Is this worth the cost and risk??
4) Although aspirin alone decreased the risk for nonfatal IHD events, there was still no difference in fatal IHD events and no difference in overall mortality -- these results are in agreement with prior studies demonstrating no overall benefit for the use of aspirin in primary prevention of IHD.
5) Other potential problems with the study -- large rate of withdrawal, loss of blinding due to minor bleeding events, limited patient population (men only), an overall incidence of IHD that was less than anticipated, and the potential bias of self selection in the initial process.
Reference: Thrombosis prevention trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischemic heart disease in men at increased risk. Lancet 1998: 351: 233-41.