DAPT’s Role After CABG Questioned: TACSI and TOP-CABG

MADRID, Spain—Adding ticagrelor on top of aspirin alone for 1 year does not lower the incidence of death, MI, stroke, or repeat revascularization in ACS patients who undergo CABG surgery, according to the TACSI trial. Moreover, this combination results in more major bleeding compared with aspirin monotherapy.

But TACSI wasn’t the only trial to take on post-CABG dual antiplatelet therapy (DAPT) at the meeting—the TOP-CABG trial suggested another route: starting out with dual ticagrelor and aspirin for 3 months, then de-escalating to aspirin alone. The idea, which bore fruit, was to maximize benefit during a vulnerable period while minimizing bleeding risk in the hopes of improving saphenous vein graft (SVG) patency.

TACSI investigator Anders Jeppsson, MD, PhD (Sahlgrenska University Hospital, Gothenburg, Sweden), presenting his group’s data in a Hot Line session at the European Society of Cardiology (ESC) Congress 2025, pointed out that both US and European revascularization guidelines recommend 12 months of DAPT with a P2Y12 inhibitor and aspirin after CABG in ACS.

Yet, these “recommendations are based on extrapolation of data from non-CABG studies, substudies of ACS trials, and smaller randomized trials with surrogate endpoints,” he said.

The findings from both TACSI and TOP-CABG were persuasive, according to the discussions that followed.

Marco Valgimigli, MD, PhD (Cardiocentro Ticino Institute, Ente Ospedaliero Cantonale, Lugano, Switzerland), who dug into the TOP-CABG trial, agreed that there’s a strong case for de-escalation.

“Thanks to TOP-CABG, I think I will never give DAPT [with] ticagrelor for 12 months to my [chronic coronary syndrome] patients after CABG,” he told the ESC audience. Valgimigli added that he may elect in selected cases—such as when SVGs are used for major coronary vessels in patients at low bleeding risk—to use ticagrelor on top of aspirin for 3 months.

TACSI

The open-label, registry-based TACSI trial, which was published simultaneously in the New England Journal of Medicine, was conducted at 22 Nordic cardiothoracic surgery centers. Researchers randomized 2,201 patients (mean age 66 years; 14.4% women) within 3 to 11 days after CABG for ACS to receive either ticagrelor (90 mg twice daily) plus aspirin (75-100 mg daily) or aspirin alone (75-160 mg daily) for 1 year after their procedure. Median baseline EuroSCORE II was 1.6%. Most (57.6%) had experienced NSTEMI, 32.1% had unstable angina, and 10.2% had STEMI.

By 1 month, 79.7% of those assigned to ticagrelor plus aspirin were still on ticagrelor, with that proportion dropping to 64.1% at 1 year. Of those assigned to aspirin, 96.5% and 92.3% still adhered to their regimen at 1 month and 1 year, respectively. Around one in 10 patients in the ticagrelor group were receiving an oral anticoagulant at 1 month and 12.8% were at 1 year. For the aspirin group, those numbers were 9.4% and 11.5%, respectively.

Adverse events leading to discontinuation of the trial drug occurred at rates of 8.9% with ticagrelor and 2.4% with aspirin alone. Nearly one in five patients in the ticagrelor group (18.2%) experienced dyspnea, as compared with 6.4% of those in the aspirin group.

At 1 year, the combined rate of death, MI, stroke, or repeat revascularization was similar for patients given ticagrelor plus aspirin and those given aspirin alone (4.8% vs 4.6%; P = 0.77). All-cause mortality was extremely rare overall (0.45%), with only four of the 10 deaths being cardiovascular. Net adverse clinical events, defined as a primary-outcome event or major bleeding, was increased with added ticagrelor (9.1% vs 6.4% with aspirin only; HR 1.45; 95% CI 1.07-1.97). This difference was driven by a doubling in major bleeding with ticagrelor compared with aspirin (4.9% vs 2.0%; HR 2.50; 95% CI 1.52-4.11).

The researchers point out that “nonadherence to ticagrelor was substantial” in TACSI, perhaps due to avoidance of triple antithrombotic therapy among patients on oral anticoagulants or because of drug side effects.

Aspirin should remain the standard of care post-CABG. There is no clear role for routine DAPT. John W. Eikelboom

John W. Eikelboom, MD (McMaster University, Hamilton, Canada), the Hot Line session’s discussant for TACSI, agreed that the trial addresses an “important and previously unresolved question” of the best antiplatelet strategy in this scenario.

For him, TACSI offers a “clear and simple” message that 12 months of ticagrelor and aspirin in CABG patients isn’t superior to aspirin alone. “The pragmatic design means that this therapy was evaluated in real-world clinical settings,” said Eikelboom, though the flip side is that the trial was underpowered for clinical events and involved a high degree of ticagrelor discontinuation.

With the addition of TACSI, the field now possesses “moderate-quality evidence” showing DAPT doesn’t decrease MACE in this population, he said. “Aspirin should remain the standard of care post-CABG. There is no clear role for routine DAPT.”

“It will be an interesting discussion,” Jeppsson said, “to see whether this [trial] changes guidelines.” He told TCTMD that he’s not yet changed his own practice, because the data are so new. It’s also necessary to see if a benefit to ticagrelor emerges over time. “We need to follow these patients a bit longer before we make any final conclusions,” said Jeppsson.

Eikelboom noted that MACE and bleeding of course aren’t the only endpoints of interest with DAPT after CABG—fully 15-20% of patients see their saphenous vein graft fail within the first year. This is important, because “graft failure predicts clinical events,” he said. High-quality evidence suggests aspirin reduces both graft failure and MACE by 50%, whereas the evidence leading up to the TACSI trial has been less consistent for DAPT.

With these new results, it’s still possible to “consider DAPT in selected patients to maintain graft patency,” he suggested.

Extended follow-up in TACSI, planned for 10 years, might provide insight into whether better patency translates into better clinical outcomes, said Eikelboom, adding that ongoing trials, such as ODIN, are set to deepen understanding of the best strategy.

TOP-CABG

The TOP-CABG trial, which was presented following TACSI by Xin Yuan, MD, PhD (Fuwai Hospital, Beijing, China), addressed the “double-edged sword” of DAPT’s ability to improve SVG patency while increasing bleeding risk. For this study, said Yuan, researchers looked into a “simple but crucial question”: whether it made sense to add ticagrelor to aspirin in the early period after CABG, when thrombotic risk is highest and then de-escalate antiplatelet therapy thereafter to aspirin alone as a way to reduce bleeding.

TOP-CABG enrolled 2,300 patients (mean age 61 years; 21% women) who underwent elective primary CABG surgery with at least one SVG at 13 centers across China. Researchers randomized them to receive ticagrelor and aspirin for 3 months, then either keep up with that same regimen or de-escalate to aspirin alone for 9 months.

We have provided a solution to the DAPT dilemma, enabling a smarter, more targeted standard of care. Xin Yuan

De-escalation, as compared with ongoing DAPT, offered noninferior SVG occlusion at 12 months (10.79% vs 11.19%; P = 0.008 for noninferiority) and less clinically relevant bleeding (8.26% vs 13.19%; P < 0.001 for superiority), particularly after 3 months.

Thus, the strategy offers a “better efficacy-safety balance,” Yuan concluded. “We have provided a solution to the DAPT dilemma, enabling a smarter, more targeted standard of care.”

Valgimigli pointed out that TOP-CABG is the largest trial yet to consider DAPT after CABG. “I think it brings clear results,” he said. “Twelve-month DAPT makes the patient bleed. Three-month DAPT is as apparently good as 12 months with [fewer] bleeding complications.”

It would be good, though, to see a trial with an aspirin-only comparator in addition to these two regimens, said Valgimigli. Moreover, it’s not definitively known whether reducing SVG will decrease hard events.

Angiographic follow-up after CABG has previously suggested “there are two components of SVG occlusion. One is immediate, periprocedure, and most likely it could be thrombotic related. Then there is another one that seems to be more linearly [distributed] over time and potentially may not be related to thrombosis,” he noted, adding that this lends support to TOP-CABG.