 |
 |
27th May 2009
The WATCH study compared adjusted dose warfarin, aspirin and clopidogrel in patients with heart failure. The study was underpowered but does not provide strong evidence that any one of these drugs is better than the others in terms of benefits or net harms.
Action
WATCH does not provide clear evidence to depart from NICE guidance on heart failure. This recommends using anticoagulants in patients with heart failure and atrial fibrillation and to consider its use in those with a history of thromboembolism, left ventricular aneurysm, or intracardiac thrombus. Aspirin is recommended for people with heart failure and atherosclerotic arterial disease (including coronary heart disease).
What is the background to this?
Many people with heart failure have a history of myocardial ischaemic events. In addition, there is a comparatively high incidence of embolic events such as stroke in people with heart failure,. These factors would suggest the need for antithrombotic therapy, but of what type? In populations without heart failure, anticoagulants are more effective than antiplatelets in preventing stroke in people with atrial fibrillation (AF), and there is some evidence for their use in people with both heart failure and AF. Optimal antithrombotic treatment in people with heart failure and sinus rhythm is unclear. Antiplatelets have a strong evidence base to support their use in ischaemic heart disease, but would the greater antithrombotic effect of anticoagulants be preferable? On the other hand, anticoagulants are associated with a higher bleeding risk, and there are theoretical reasons to suspect an interaction between aspirin and ACE-inhibitors to reduce the latter’s effect. WATCH was designed to determine the optimal antithrombotic drug in people with heart failure in sinus rhythm. It compared open-label warfarin (target INR 2.5 to 3.0) with double blind aspirin 162mg/day or clopidogrel 75mg/day.
What does this study claim?
WATCH was underpowered: slow enrolment led to it being stopped prematurely, with only 1587 patients enrolled (a little over a third of the planned number) and a mean duration of follow-up of only 1.9 years instead of the planned 3.5 years. This limits the conclusions that can be drawn from it. There were no statistically significant differences between the groups in relation to the primary outcome, a composite of all-cause mortality, nonfatal myocardial infarction (MI) or nonfatal stroke. The hazard ratio for the warfarin-aspirin comparison was 0.98 (95% confidence interval [CI] 0.86 to 1.12, P=0.77), for the clopidogrel-aspirin comparison it was 1.08 (95%CI 0.83 to 1.40, P=0.57) and for the warfarin-clopidogrel comparison it was 0.89 (95%CI 0.68 to 1.16, P=0.39). The incidence of stroke was 2.3% in both antiplatelet groups, compared to 0.6% in the warfarin group (relative risk [RR] 3.83, P=0.016, NNH=59). In the aspirin group, 22.2% of patients were admitted to hospital at least once with decompensated heart failure, compared with 18.5% in the clopidogrel group and 16.5% in the warfarin group. The difference between aspirin and warfarin was statistically significant (P=0.019) but the differences between aspirin and clopidogrel, and clopidogrel and warfarin were not (P=0.14 and P=0.38, respectively). However, when time to heart failure admission or death was considered, there were no significant differences between the groups. Major haemorrhage was more common in the warfarin group (5.2%) than the aspirin group (3.6%) and the clopidogrel group (2.1%). The difference between the clopidogrel and warfarin groups was statistically significant (P=0.007) but the differences between aspirin and clopidogrel, and aspirin and warfarin were not (P=0.14 and 0.22, respectively).
So what?
Interpretation of WATCH is seriously hampered by its lack of power. There does not seem to be much to choose between any of the drugs to reduce the risk of the composite primary endpoint. However, important questions regarding safety remain unanswered.
The difference in admission rates for heart failure between aspirin and warfarin appears to give support to the theory that aspirin has a negative effect in this regard. However, observed differences in secondary endpoints must be viewed with caution when the primary endpoint does not achieve statistical significance. The P value indicates a 1 in 50 chance that the observed difference was due to chance. An adverse effect of aspirin in heart failure has been predicted because of a proposed interaction with ACE-inhibitors due to aspirin’s effects on prostaglandins: clopidogrel does not share this (theoretical) interaction, yet admission rates for heart failure in both antiplatelet groups were similar. This illustrates the need for good patient outcome data wherever possible instead of providing treatment based solely on pharmacological models. Similarly, it is difficult to draw firm conclusions with regard to comparative bleeding risks.
This study does not seem to provide firm evidence to depart from NICE guidance.
More information on heart failure and its management can be found on the heart failure section of NPC
Study details
Design Randomised controlled trial: one arm open-label, other two arms double blind
Patients: 1587 patients with symptomatic heart failure (NYHA class II to IV, 98% II and III) for at least 3 months, in sinus rhythm. Mean age was 63 years. 58% had had prior MI, 99% were taking lop diuretics, 87% were taking an ACE-inhibitor, 10% were taking an ARB, 70% were taking a beta-blocker, 51% were taking digoxin and 28% were taking spironolcatone.
Intervention and comparison open-label warfarin (target INR 2.5 to 3.0) with double blind aspirin 162mg/day or clopidogrel 75mg/day, for a mean of 1.9 years.
Outcomes and results. There were no statistically significant differences between the groups in relation to the primary outcome, a composite of all-cause mortality, nonfatal myocardial infarction (MI) or nonfatal stroke. The hazard ratio for the warfarin-aspirin comparison was 0.98 (95% confidence interval [CI] 0.86 to 1.12, P=0.77), for the clopidogrel-aspirin comparison it was 1.08 (95%CI 0.83 to 1.40, P=0.57) and for the warfarin-clopidogrel comparison it was 0.89 (95%CI 0.68 to 1.16, P=0.39). The incidence of stroke was 2.3% in both antiplatelet groups, compared to 0.6% in the warfarin group (relative risk [RR] 3.83, P=0.016, NNH=59). In the aspirin group, 22.2% of patients were admitted to hospital at least once with decompensated heart failure, compared with 18.5% in the clopidogrel group and 16.5% in the warfarin group. The difference between aspirin and warfarin was statistically significant (P=0.019) but the differences between aspirin and clopidogrel, and clopidogrel and warfarin were not (P=0.14 and P=0.38, respectively). However, when time to heart failure admission or death was considered, there were no significant differences between the groups. Major haemorrhage was more common in the warfarin group (5.2%) than the aspirin group (3.6%) and the clopidogrel group (2.1%). The difference between the clopidogrel and warfarin groups was statistically significant (P=0.007) but the differences between aspirin and clopidogrel, and aspirin and warfarin were not (P=0.14 and 0.22, respectively).
Sponsorship: Supported and administered by the Cooperative Studies Program of the Office of Research and Development, Department of Veterans Affairs. Study drugs and unrestricted grants were provided by Bristol Myers Squibb, Sanofi-Synthelabo, and Dupont pharmaceutical companies.
Feedback
Please comment on this blog in the NPC discussion rooms, or using our feedback form.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.