Paradoxical myocardial infarction risk with dual antiplatelet therapy in East Asian ischaemic stroke: a nationwide real-world cohort study and pharmacogenomic risk prediction model.

BACKGROUND

Antithrombotic guidelines for secondary stroke prevention are derived predominantly from Western populations, yet East Asian patients exhibit markedly higher cytochrome P450 2C19 (CYP2C19) loss-of-function allele prevalence (50%-60% vs ~25% in Europeans) - polymorphisms essential for clopidogrel bioactivation - potentially generating population-specific cardiovascular risks not captured in Western-derived recommendations.

OBJECTIVES

To characterise the 100-day cardiovascular safety and effectiveness of antithrombotic regimens in East Asian ischaemic stroke patients and to develop a pharmacogenomically informed clinical risk prediction model.

DESIGN

Nationwide retrospective cohort study.

METHODS

We analysed 5746 adults from Taiwan's National Health Insurance Research Database (2011-2020) stratified by antithrombotic regimen initiated within 48 h of confirmatory neuroimaging. Primary outcomes - recurrent ischaemic stroke, haemorrhagic stroke, acute myocardial infarction, heart failure and all-cause mortality - were assessed over 100 days. Multivariable Cox regression, propensity score matching, inverse probability weighting, Fine-Gray competing risks models, E-value bias quantification and negative control outcome analysis were applied. The East Asian Post-Stroke Antithrombotic Risk (EPSAR) score was developed incorporating age, sex, comorbidities and treatment regimen.

RESULTS

Dual antiplatelet therapy (DAPT; aspirin plus clopidogrel) was associated with a 17.2-fold increased acute myocardial infarction risk versus aspirin monotherapy (95% confidence interval (CI) 5.61-52.7; number needed to harm 14). This signal was robust across all analytical approaches (propensity-matched hazard ratio (HR) 11.8; inverse probability of treatment weighting HR 13.4; competing-risks subdistribution HR 14.8) and yielded an E-value of 33.4, substantially exceeding the strength of known clinical confounders. Warfarin-containing regimens conferred markedly elevated haemorrhagic stroke risk (HR 7.11-8.77). Aspirin monotherapy demonstrated a 94% all-cause mortality reduction versus no treatment. Negative control outcomes showed no systematic bias (joint test p = 0.98). The EPSAR model achieved a C-statistic of 0.78 (95%CI 0.72-0.84).

CONCLUSION

DAPT is associated with unexpected myocardial infarction risk in East Asian stroke patients, consistent with CYP2C19-mediated pharmacogenomic vulnerability. These findings challenge the universal applicability of Western-derived guidelines and support integration of pharmacogenomic risk stratification into antithrombotic prescribing for East Asian populations.

CLINICAL TRIAL REGISTRATION: Not applicable (observational study).