Smoking and 10-year risk of cardiovascular and non-cardiovascular events after contemporary coronary stenting.

The impact of smoking cessation on long-term clinical outcomes after contemporary percutaneous coronary intervention (PCI) is not well known. We estimated the association of smoking and smoking cessation on the 10-year risk of cardiovascular and non-cardiovascular events in patients after contemporary PCI in a multicenter retrospective cohort of all patients having PCI with second generation drug-eluting stents in the VA Healthcare System between 2008 and 2016. Smoking status, comorbidities and clinical outcomes were extracted from the medical record and the National Death Index. Hazard ratios and 95 % confidence intervals (HR, 95 %CI) were derived from Cox Proportional Hazards Models. Estimates of events prevented from smoking cessation were derived from the differences in absolute risks between subjects who smoked and had stopped smoking. Among 29,001 patients, 10,598 (36.5 %) were current smoking patients, 13,093 (45.1 %) were former smoking patients, and 5,310 (18.3 %) never smoked. Over 10 years, 7,806 (26.9 %) subjects died with non-cardiovascular deaths exceeding cardiovascular deaths. In multivariable models, current smoking was significantly associated with increased long-term risks of all-cause death (HR = 1.27, 95 %CI = 1.19, 1.36), myocardial infarction (HR = 1.32, 95 %CI = 1.21-1.43), cancer death (HR = 2.55, 95 % CI = 2.10, 3.08), and pulmonary death (HR = 4.07, 95 % CI = 2.85, 5.83). Smoking cessation may prevent 18.5 % (95 %CI = 16.0 %, 20.9 %) all-cause deaths, 14.8 % (95 %CI = 9.8 %, 19.6 %) cardiovascular deaths, 42.6 % (95 %CI = 37.7 %, 47.2 %) cancer deaths, and 48.3 % (95 %CI = 41.8 %, 54.2 %) pulmonary deaths among smokers. Stopping smoking will likely have major impacts on non-cardiovascular events as well as cardiovascular events in patients after PCI with second generation drug-eluting stents.

Déclaration de conflit d'intérêts

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.