Background and purpose Studies suggest that smoking affects clopidogrel efficacy. However, whether it influences the association between CYP2C19 genetic variants and clopidogrel efficacy is not clear. Methods In total, 2961 patients from the CHANCE trial were involved in this substudy and were successfully genotyped for two single-nucleotide polymorphisms of CYP2C19 (*2 and *3). The Cox proportional risk regression model was used to evaluate the interactions between CYP2C19*2 and CYP2C19*3 carrier status and clopidogrel efficacy stratified by smoking status. Results There were marginal significant interactions between CYP2C19*2 and CYP2C19*3 allele carrier status and antiplatelet treatment regimen for the risk of recurrent stroke and composite events (P = 0.054, P = 0.051, respectively) amongst smokers, but not in non-smokers. Amongst smokers, clopidogrel plus aspirin decreased the recurrence rate of stroke compared with aspirin alone in non-carriers (3.8% vs. 11.8%, hazard ratio 0.32, 95% confidence interval 0.15-0.65, P = 0.002), but not in carriers. Similar results were also found for the recurrence rate of composite events in smokers. No significant difference was found for hemorrhage events in any group. Conclusions Amongst patients with minor stroke or transient ischaemic attack, marginal significant interactions between CYP2C19*2 and CYP2C19*3 allele carrier status and clopidogrel efficacy were found in smokers but not in non-smokers. Amongst smokers, clopidogrel plus aspirin might decrease the recurrence rate of stroke in non-carriers of *2 and *3 alleles of CYP2C19 compared with aspirin alone. However, caution should be taken to interpret our findings in view of several limitations in our study.