May 08, 2011

Aspirin in Primary Prevention: New Meta-Analysis

From Heartwire
April 29, 2011 (Birmingham, Alabama) -- A new meta-analysis of studies of aspirin in primary prevention in a total of 90 000 subjects has suggested a 14% reduction in total cardiovascular events (driven by a 19% reduction in nonfatal myocardial infarction [MI]) and a nonsignificant reduction in overall mortality and stroke.

But clinical-trials guru Dr Sanjay Kaul (Cedars Sinai Medical Center, Los Angeles, CA) is not overly impressed with these new data and continues to be skeptical about the role of aspirin in primary prevention.

Kaul commented to heartwire : "The key finding of this meta-analysis is that aspirin use is associated with a statistically significant reduction in nonfatal MI. However, the clinical significance of this finding is not clear, as annualized risk difference or the number-needed-to-treat data are not presented. It is also not clear whether the data analysis included silent MIs identified on ECG examination. It is important to emphasize that, with the exception of one trial (the Thrombosis Prevention Trial), the primary end point was not met in any one of the studies included. For guiding clinical practice, an estimate of benefit/risk is necessary, which these data don't provide. In addition to significant statistical heterogeneity, there is also clinical heterogeneity (in aspirin dose, treatment duration, concomitant medications, etc) that might impact interpretation of the data. Lack of patient-level data precludes adjustment for variability in critical prognostic covariates and does not permit the more robust time-to-event analysis."

The latest meta-analysis, published online April 8, 2011 in the American Journal of Cardiology, adds three new studies--the Aspirin for Asymptomatic Atherosclerosis (AAA) trial, the Prevention of Progression of Arterial Disease and Diabetes (POPADAD) trial, and the Japanese Primary Prevention of Atherosclerosis With Aspirin for Diabetes (JPAD) trial--to the previous meta-analysis published by the Antithrombotic Trialists' Collaboration (ATTC) in 2009.

Lead author Dr Alfred Bartolucci (University of Alabama at Birmingham) told heartwire : "Our results are in line with the ATTC meta-analysis, but we have added more patients, so the results become stronger."

Meta-Analysis: Predefined End Points

End pointOdds ratio (95% CI)p
Total CHD0.85 (0.69–1.06)0.154
Nonfatal MI0.81 (0.67–0.99)0.042
Total CV events0.86 (0.80–0.93)0.001
Stroke0.92 (0.83–1.02)0.116
CV mortality0.96 (0.80–1.14)0.619
All-cause mortality0.94 (0.88–1.01)0.115
But What About the Bleeding Risk?

Bartolucci was reluctant to be drawn into the controversy that has been simmering recently over claims that recommendations for the use of aspirin in primary prevention are too enthusiastic. This has been based on concerns about the bleeding risk associated with taking daily aspirin. In particular, the ATTC investigators found that the same people who would derive the greatest benefit (those at higher risk of heart disease) are also at higher bleeding risk with aspirin. This led them to conclude that "there is not good evidence of substantial benefit that outweighs risk enough to justify a public policy recommending routine use in primary prevention."

The current meta-analysis did not show a significant increase in gastrointestinal (GI) bleeding with aspirin, but Bartolucci said he would not recommend that people take aspirin for primary prevention if they were at risk of GI bleeding.
The GI bleeding rate with aspirin varied in the nine studies included from 0.3% to 4.5%. Bartolucci attributed this variation to different patient populations.

He noted that many of these studies included people who were not at especially high risk of heart disease. "Many of these studies just included normal middle-aged individuals. Some had risk factors, some did not."

Bartolucci was not eager to make recommendations based on his data. "We're just reporting the data. People have to interpret it for themselves," he told heartwire. "Our bottom-line results apply to the average person. But not everyone fits into the average."

When pushed, he said he thought there was a major role for aspirin in primary prevention but that people needed to consult with their doctor to make sure they could tolerate it. He added: "The benefit of aspirin in primary prevention is still there. And we are now getting more knowledge of how different groups of individuals, such as diabetics, are at risk of CHD."
But Kaul disagreed. "In my opinion, the proper use of a meta-analysis is not only to harness power from inadequately powered individual trials to derive a pooled estimate of treatment effect but also to identify consistency of treatment effects across important subgroups. I also do not consider a meta-analytic p value of <0.05 to provide strong evidence," he told heartwire .

Kaul added: "Positive secondary end points have typically (and generously) been used to support aspirin recommendations by professional societies. However, the FDA continues (rightly so) not to endorse aspirin for primary prevention, even without considering the contemporary negative trials."

This study was supported by an unrestricted research grant from Bayer HealthCare AG.

No comments:

Post a comment

Got something to say? We appreciate your comments: