Background Thrombus aspiration is a mechanistically logical adjunct in primary angioplasty for acute myocardial infarction. Individual randomised controlled trials (RCTs) have not shown a consensus of mortality reduction and there are concerns about stroke. We perform a meta-analysis of all available RCT data on thrombus aspiration, including the large, recently published TOTAL trial.

Method and results A meta-analysis of RCTs of thrombus aspiration, including the recent TOTAL data was performed. At 30 days (11 trials; 20,192 patients) there was a marginally non-significant reduction in all cause-mortality with thrombus aspiration (OR 0.844, 95% CI 0.710–1.003, p = 0.05). The longer-term follow-up data (13 trials; 20,142 patients) was similarly non-significant (OR 0.89, 95% CI 0.78–1.01, p = 0.08). At both 30 days and longer-term follow-up, there was a statistically significant increase in stroke with thrombus aspiration (OR 1.56, 95% CI 1.05 to 2.32, p = 0.03, and OR 1.94, 95% CI 1.24 to 3.04, p = 0.04 respectively).

Conclusions The point estimates in the meta-analyses suggest that thrombus aspiration may prevent four deaths per thousand at the cost of two strokes per thousand. Although this may initially sound favourable for the procedure, the confidence interval for mortality is still wide enough to encompass no effect, while that of stroke is not.

With mortality now so low in STEMI trials, very large numbers of patients are required to reliably identify a clinically important improvement. The task requires massive multi-centre trials or strategies that minimise per-patient costs by using established outcome-reporting infrastructure to focus on mortality, and perhaps introduce “retrospective consent”.