Trials assessing the value of adjunctive aspiration thrombectomy before TASTE were much smaller, a little inconsistent and had more focal endpoints, such as myocardial blush. TASTE instead looked at a hard clinical endpoint of 30-day all-cause mortality. The endpoint wasn’t reached in TASTE, indicating that there was no clinical benefit in performing aspiration thrombectomy. With the TAPAS trial, at 1 year they did show a mortality benefit, so the question is why TASTE did not. Although I can’t definitively answer that question, I can say that this only looked at 1 month data, and perhaps positive effects will manifest at 1 year. 

This was a large trial of several thousand patients with STEMI undergoing primary PCI in Sweden. The idea was to see whether mortality would be favorably influenced, the hypothesis being perhaps removal of a thrombus before stenting would reduce mortality. There had been meta-analyses — in fact, I was a senior author for two of them — that showed a lower rate of major adverse cardiac events and even a lower rate for mortality [with this technique] in the 6- to 12-month range. As it turns out, this large, well-powered study did not show any significant difference in mortality, so manual thrombus aspiration did not reduce mortality. I should point out that it was a 30-day endpoint that was presented here, so we really need to see what happens with longer-term follow-up, and such follow-up is planned in this registry at 1 year, 3 years and 5 years. We will have a better sense, with longer-term follow-up, whether there is or is not a reduction in mortality. Secondary endpoints such as MI were reduced, although not statistically significantly so, just trends in terms of other secondary endpoints. Some practicality endpoints were reduced; the ability to direct stent was improved, the stent length was reduced with the strategy of manual thrombus aspiration before stenting. 

This is a simple procedure; it is relatively inexpensive and doesn’t add much time to the procedure. I probably will continue manual thrombus aspiration in patients who present with STEMI prior to stenting them. On the other hand, one has to acknowledge that this large, well-done trial did not find a mortality difference. I should mention, however, there is another large randomized clinical trial that is ongoing, the TOTAL trial. We should wait and see what the TOTAL trial shows and what longer-term follow-up from the TASTE trial shows before making any decisions regarding whether to abandon thrombus aspiration using a catheter.

What we know from previous clinical trials is that thrombus aspiration tends to reduce distal embolization of thrombus and prevents low reflow. And clinically, we find that as an interventionalist, it’s quite useful in patients with heavy thrombus, for example. And we know from data from the American College of Cardiology – National Cardiovascular Data Registry that in about 20% of STEMIs, interventionalists use thrombus aspiration.

The TASTE trial has a very ambitious design, in that it’s the first trial probably in the last 5 to 10 years to use all-cause mortality as a primary outcome. And the reason why that’s unique is because mortality rates in STEMI are so low that [sponsors of] most trials, particularly [those funded by] pharmaceutical and device companies, have decided that it simply isn’t economically feasible to do a trial for all-cause mortality. When I first heard about this design, I was very happy that they were trying to go for such an ambitious outcome, but I was also concerned when I read the design paper that they initially planned for 10% control-rate mortality, because 10% mortality in a randomized clinical trial, even in an all-comers registry-based trial, is pretty unlikely. And they had data that they presented at various meetings on the mortality rates that they were observing, and they were somewhere between 2% to 3%. So you realize all of the sudden, if a trial was powered for 10% mortality, then going from 5% to 7% is not going to do the job in terms of accounting for that reduction in control event rates.