BIOANALYSISPAPER
A patient level meta analysis of six trials and 10,634 patients finds the bleeding win came from removing aspirin, with a trade off in the first 14 days.
After a patient with atrial fibrillation lands in the cath lab and walks out with a heart stent, the prescribing question arrives immediately: which blood-thinning regimen goes on the discharge sheet? The old default, warfarin plus aspirin plus a second platelet-blocker (dual antiplatelet therapy, or DAPT), was built to keep both the stent and the irregular heart rhythm from throwing clots. A new patient-level meta-analysis of six randomized trials and 10,634 patients, published in the European Heart Journal and registered with PROSPERO, says the bleeding cost of that regimen was largely the second antiplatelet. Drop it, and TIMI major bleeding roughly halves, with no statistically significant difference in the one-year composite of cardiovascular death, myocardial infarction, or stroke (PMID 42669062).
The four-arm comparison is what makes the analysis decision-relevant. The trials randomized AF patients undergoing PCI to direct oral anticoagulant (DOAC) plus a single P2Y12 inhibitor, vitamin K antagonist (VKA, the warfarin class) plus single antiplatelet, VKA plus DAPT, or DOAC plus DAPT. Arm sizes ran 4,083, 1,247, 3,715, and 1,589 respectively. Against the VKA-plus-DAPT default, the DOAC-plus-P2Y12 regimen cut TIMI major bleeding to a hazard ratio of 0.48 (95% CI 0.38 to 0.65), and beat VKA-plus-single-antiplatelet at 0.62 (0.40 to 0.97). Bleeding reductions against DOAC-plus-DAPT were borderline. The result is consistent with what the contributing trials each suggested on their own: the largest modifiable bleeding driver is the aspirin layer rather than the anticoagulant class (Vivli registry).
The trade-off lives in the first two weeks. A 14-day landmark analysis of the same pooled cohort found higher rates of myocardial infarction and definite or probable stent thrombosis in patients on the shorter-DAPT regimens, DOAC plus a single P2Y12 inhibitor, and VKA plus single antiplatelet, during the early post-PCI window. In the DOAC-plus-P2Y12 arm, the median transition from DAPT to single antiplatelet was one day (interquartile range 1 to 3); in the VKA-plus-single-antiplatelet arm, it was three days (IQR 1 to 7). One-year composites did not differ significantly across arms, with DOAC plus P2Y12 versus VKA plus DAPT at 1.16 (0.97 to 1.41), VKA plus single antiplatelet at 1.14 (0.85 to 1.54), and DOAC plus DAPT at 0.99 (0.75 to 1.30), but the early-period signal is the open question the next trial generation will need to design around. No significant interaction with age, sex, bleeding risk, or thrombotic risk was detected at one year (PMID 42669062).
The clinical implication is narrow but specific. For the patient with AF and no recent acute coronary syndrome, the evidence now supports a one-year regimen that drops aspirin early and keeps an oral anticoagulant paired with a single platelet-blocker, the regimen with the cleanest bleeding profile and a non-significant difference in ischemic events at twelve months. For the first two weeks, the residual MI and stent-thrombosis signal argues against a literal day-one drop of DAPT across the board; trial-level operationalization of the transition is where the next set of studies will need to land. The meta-analysis is not a society guideline, and the contributing trials already span the decade in which DOAC practice moved from novel to standard. The reading that survives scrutiny is this: the bleeding win came from removing aspirin, the one-year ischemic composite is statistically non-significant rather than proven equivalent, and the early-period signal is part of the answer, not an asterisk (PMID 42669062).