In a four-hospital Chinese observational cohort, post-PCI anticoagulant therapy, or PACT, showed no statistically significant association with either in-hospital death or major bleeding among patients with high thrombotic risk and STEMI. Among the 2,001 patients included, 2.1% of those who received PACT died in hospital, compared with 2.3% of those who did not. Major bleeding occurred in 11.8% of the PACT group and 13.5% of the non-PACT group. The reported P values were 0.821 for death and 0.637 for major bleeding.
That is a finding about association, not proof that PACT has no benefit or no bleeding risk. Because the study was observational and clinicians selected treatment details, the groups could differ in ways the analysis could not fully measure. The authors caution that the results do not establish equivalence or non-inferiority between PACT and no PACT.
How the groups were formed
Researchers used medical-record data for 5,335 STEMI patients who underwent PCI at four Chinese hospitals between 2010 and 2022. High thrombotic risk was classified using TWILIGHT criteria, requiring at least one clinical feature and one angiographic feature. Examples included age 65 or older, an eGFR below 60 mL/min/1.73 m², and total stent length above 30 mm.
Of the 5,335 enrolled patients, 2,001 were included. The PACT group contained 473 patients, or 23.64% of the analysis, while 1,528 patients, or 76.36%, did not receive PACT. Patients were grouped by treatment received, while clinicians chose the anticoagulant type and duration at their discretion.
The primary outcomes were all-cause death and major bleeding during hospitalization. Secondary outcomes included MACEs, NACEs, any bleeding and a post-discharge outcome combining death and stroke. Major bleeding used BARC grades 3 through 5 and included a hemoglobin decrease of at least 4 g/dL among its criteria; any bleeding included a drop of at least 3 g/dL.
The statistical checks stayed inconclusive
After multivariable adjustment, PACT was not significantly associated with in-hospital death. The adjusted odds ratio was 0.70, with a 95% confidence interval from 0.33 to 1.48 and P = 0.349. The adjusted result for major bleeding was 0.84, with a 95% confidence interval from 0.32 to 2.18 and P = 0.720. The authors say the wide intervals mean the analysis does not establish equivalence or non-inferiority.
A time-varying Cox analysis gave hazard ratios of 0.65 for in-hospital death and 0.62 for major bleeding. Their reported 95% confidence intervals were 0.30 to 1.41 and 0.10 to 3.74, respectively, with P values of 0.274 and 0.604. The estimates pointed in the same direction as the main analysis and did not materially change the inference.
Researchers also used 1:1 propensity matching to balance measured characteristics. It left 171 PACT patients and 171 non-PACT patients. In that matched analysis, the odds ratio was 0.56 for death and 1.51 for major bleeding. The 95% confidence intervals were 0.16 to 1.95 and 0.25 to 9.15, and the reported P values were 0.364 and 0.655, so neither result was statistically significant.
At a median follow-up of 1.72 years, adjusted hazards were also similar: 0.94 for all-cause death and 1.03 for MACEs. The reported P values were 0.747 and 0.873, and the interquartile range for follow-up was 0.04 to 3.05 years.
Across age, sex, diabetes, hypertension and renal-insufficiency subgroups, the study reported no significant interaction or heterogeneity in the in-hospital associations.
A cautious result, not a verdict
The limitations matter because treatment was selected in clinical practice. The authors identified possible selection bias, residual confounding and physician-dependent treatment and follow-up. They also lacked data on sheath size and closure technique. Matching reduced the sample to 171 pairs, limiting precision, and the findings have limited applicability to patients excluded because they had anticoagulation indications.
Taken together, the authors concluded that routine post-PCI PACT was not significantly associated with lower in-hospital mortality or ischemic events, or with higher major bleeding. They cautioned that the observational design and wide confidence intervals do not establish equivalence or non-inferiority. The result describes what was observed in this selected high-risk STEMI cohort, not a definitive answer about the best treatment for every patient.
Paper data and sources
Original title: Association between post procedural parenteral anticoagulation therapy and clinical outcomes in ST-elevation myocardial infarction patients with high thrombotic risk
Authors: Lai P, Dai Y, Yin M et al.
Journal/Repository: Frontiers in pharmacology
Status: Peer-reviewed
First online: 2026-10-07
DOI: Not available
Original paper · Full text