Table 3.
—[Sections 3.1.1-3.1.5, 3.2.1] Aspirin vs No Aspirin in Patients With Established CAD9
| Outcomes | Participants (Studies), Follow-up | Quality of the Evidence (GRADE) | Relative Effect (95% CI) | Anticipated Absolute Effects Over
5 y |
|
| Risk Without Aspirin | Risk Difference With Aspirin (95% CI) | ||||
| Total mortality |
17,000 (16 RCTs), 27 mo |
Moderate due to imprecisiona |
RR 0.90 (0.82-0.99) |
133 per 1,000b |
13 fewer per 1,000 (from 24 fewer to 1
fewer) |
| MI nonfatal events |
17,000 (16 RCTs), 27 mo |
High |
RR 0.69 (0.60-0.80) |
117 per 1,000b |
37 fewer per 1,000 (from 47 fewer to 23
fewer) |
| Stroke includes nonfatal ischemic and hemorrhagic
strokesc |
17,000 (16 RCTs), 27 mo |
High |
RR 0.81 (0.71-0.92) |
135 per 1,000b |
26 fewer per 1,000 (from 39 fewer to 11
fewer) |
| Major extracranial bleed | 17,000 (16 RCTs), 27 mo | Moderate due to indirectnessd | RR 2.69 (1.25-5.76) | 15 per 1,000e | 25 more per 1,000 (from 4 more to 71 more) |
CAD = coronary artery disease; CAPRIE = Clopidogrel vs Aspirin in Patients at Risk of Ischemic Events; CHARISMA = Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance; RCT = randomized controlled trial. See Table 1 and 2 legends for expansion of other abbreviations.
Rated down for imprecision because the 95% CI suggests possible benefit and no effect on total mortality.
Control group risk estimates (without aspirin) for MI and stroke come from observed yearly event rates in 16 RCTs reported in the meta-analysis, adjusted to a 5-y time frame. The control group rate estimate for total mortality without aspirin is derived from the event rate in the aspirin arm of the CHARISMA trial, using the RR of 0.90 to get the control group rate estimate without aspirin.
Of the strokes in the meta-analysis, 0.8% with aspirin were intracranial hemorrhages, and 0.4% of strokes without aspirin were intracranial hemorrhages.
Rated down for indirectness because bleeding events were only reported in a subset of trials with stroke and transient ischemic attack populations.
To estimate control group risks for major bleeds, we have used major bleed event rates from the aspirin arm in the CAPRIE trial adjusted to a 5-y time frame as the starting point (to ensure consistency across evidence profiles). We then used the RR of 2.69 for the comparison of aspirin to no aspirin observed in the meta-analysis to derive the control group rate estimate without aspirin.