Secondary prevention with aspirin [February 2003; 108-3]

Secondary prevention with aspirin
Systematic review Results Comment

The Antithrombotic Trialists' Collaboration was a remarkable tour de force. The trouble some would see is that in combining 195 trials with different antithrombotics in different patients for different times the analysis misses the main point: that of convincing physicians to prescribe aspirin in low doses for secondary prevention.

This has been addressed in a different analysis [1]. Here the essence was to find trials that informed on the FDA-approved uses of low dose (50-325 mg/day) aspirin. These are in the secondary prevention of myocardial infarction and stroke in patients with previous infarction, stroke or transient ischaemic attack, or those patients with a history of angina.

Systematic review

Searching used electronic databases for all published reports since 1970 that informed on the FDA's updated professional labelling for aspirin. For inclusion, studies had to be randomised, with a placebo control, and use aspirin at doses of 50-325 mg/day. Excluded were trials of duration shorter than three months, trials that used short-term prophylaxis in procedures (like angioplasty or bypass surgery), trials that combined aspirin with another agent or used aspirin in otherwise healthy individuals.

Outcome data collected included myocardial infarction, stroke, vascular death, vascular events (any stroke, myocardial infarction or vascular death), and all cause mortality. Gastrointestinal bleeding events were included regardless of severity.

Results

There were six trials with 6,300 patients; about 4,200 had had a myocardial infarction or stroke. Subjects were predominantly men, with mean age in trials of mid-50s to mid-60s. Duration of studies appeared to be from three to 52 months.

All six trials had fewer deaths with aspirin than with control. There were significantly fewer deaths, combined vascular events, and myocardial infarctions, but more gastrointestinal bleeds (Table 1). Of the 41 gastrointestinal bleeds with aspirin 24 were severe, and of the 17 bleeds with placebo 9 were severe. No deaths were caused by bleeding.

Table 1: Outcomes in secondary prevention trials of aspirin

	Number of events		Percent of events
Aspirin	Aspirin	Placebo	Aspirin	Placebo	Relative risk (95% CI)	NNT/H (95% CI)
Total number	3127	3173
Deaths	241	291	7.7	9.2	0.8 (0.7 to 0.99)	68 (35 to 1080)
All vascular events	607	788	19.4	24.8	0.7 (0.6 to 0.8)	18 (13 to 30)
Myocardial infarction	234	324	7.5	10.2	0.7 (0.6 to 0.8)	37 (25 to 75)
Stroke	193	231	6.2	7.3	0.8 (0.7 to 1.0)	90 (43 to 780)
GI bleeding	41	17	1.3	0.5	2.5 (1.4 to 4.7)	129 (80 to 330)

Benefits outweighed risks (Figure 1). The results indicated that if 1000 patients were treated with aspirin prophylactically as secondary prevention instead of not being treated, there would be 55 fewer vascular events (including 27 myocardial infarctions and 11 strokes), and 15 fewer deaths. Off set against this benefit would be 8 more episodes of bleeding, of which about half would be severe.

Figure 1: Crude pooled event rates of outcomes from six trials

Comment

This is reassuring information for those who want to know that aspirin is effective and safe not across every high risk eventuality, but for the patient who they treat most frequently. It is also very reassuring for patients. The missing information is the weighted duration across all six studies.

Reference:

SM Weisman, DY Graham. Evaluation of the benefits and risks of low-dose aspirin in the secondary prevention of cardiovascular and cerebrovascular events. Archives of Internal Medicine 2002 162: 2197-2202.

previous or next story in this issue