This cross-sectional study of community-dwelling Korean men demonstrated that cumulative smoking exposure was positively associated with PAD, defined as an ABI <0.90, and that duration of smoking cessation was inversely associated with PAD, independently of conventional cardiovascular risk factors. Although, as in previous studies, current smokers had a significantly higher risk for PAD than those who never or previously smoked, pack-years of smoking and years since quitting smoking were also identified as significant risk factors that should be taken into account to properly evaluate the effect of cigarette smoking on PAD.
The ABI is considered as a surrogate marker of generalized atherosclerosis because low ABI levels have been associated with elevated risk of future coronary heart disease [3, 22], stroke [3, 23, 24], and a higher risk of all-cause and cardiovascular mortality [3, 25, 26]. A recent meta-analysis found that the predictive value of ABI for cardiovascular morbidity and mortality was similar to that of the traditional Framingham risk factors . Because ABI is simple, non-invasive, reproducible, and cost-effective , ABI measurement is routinely performed to screen for asymptomatic PAD patients. ABI is a useful tool for the prediction of cardiovascular risk because, compared with carotid intima-media thickness and coronary artery calcium, it can be performed readily in community settings and in primary care physicians' offices .
Our results are consistent with previous findings that current smoking is associated with a 1.86 to 5.48-fold increase in risk of PAD, compared with never having smoked [7–14]. Some studies have reported that current smoking is associated with a 1.60 to 2.80-fold increase in the risk of PAD, compared with never or previously smoking [29, 30]. Additionally, our findings are analogous to previous findings that former smoking is associated with a 1.02 to 1.94-fold increased risk of PAD [7, 9–11, 14]; however, generally no significant difference between former and never smoking has been found.
Although various epidemiological studies have reported an association between smoking status and PAD, the majority of studies have examined PAD risk by smoking status, finding that current smoking increased PAD risk or smoking cessation decreased PAD risk. However, the simple smoking status classification of never, former, and current smoking has been questioned on the assumption that lifetime heavy smokers who have recently stopped smoking are categorized as former smokers, whereas smokers who started smoking only a few months ago are categorized as current smokers. Thus, cumulative smoking exposure, such as pack-years, should be taken into account to properly evaluate the association between cigarette smoking and PAD. Because few studies have investigated the association between cumulative smoking exposure and PAD or between the duration of smoking cessation and PAD, the dose-response relationship between smoking habits (e.g., pack-years of smoking and years since quitting smoking) and PAD is uncertain. The Edinburgh Artery Study of 1592 adults reported that the adjusted relative risk of PAD, compared with never smoking, was 1.70 (95% CI, 0.72-3.99) for moderate smokers (≤25 pack-years) and 2.72 (95% CI, 1.13-6.53) for heavy smokers (> 25 pack-years) . Recently, a cross-sectional study of 1215 Japanese men found that mean ABI correlated inversely and linearly with pack-years of smoking and the OR (95% CI) of PAD, compared with never smoking, was 2.8 (0.8-10.2), 2.8 (0.8-10.0), and 4.2 (1.2-14.6) for <26, 26-45, and ≥45 pack-years, respectively, suggesting a linear trend . In agreement with previous studies [13, 15], we observed a significant increasing trend between pack-years of smoking and PAD, confirming that the effect of cumulative smoking exposure on PAD was a dose-response relationship. Cui et al.  found that men who had quit smoking 20 or more years ago had higher mean ABI and lower prevalence of PAD than current smokers. In our study, the risk of PAD was significantly lower with smoking cessation of over 11 years than in current smokers, whereas ≤10 years of smoking cessation was not significantly associated with PAD risk, suggesting that long-term smoking cessation is needed to diminish the effects of smoking on PAD.
Study limitations and strengths
There are several limitations to the present study. First, the cross-sectional nature limits conclusions about the direction or causality of associations observed in our study. Additional prospective studies with incident PAD are needed to confirm our findings. Second, the possible measurement errors in smoking characteristics due to the information being self-reported, without measuring biological markers, such as serum cotinine, might have attenuated the relationship between cigarette smoking and PAD. Third, unmeasured confounding variables could have affected the association of cigarette smoking and smoking cessation with PAD. Fourth, because information on environmental tobacco smoke exposure was not collected, the effect of passive smoking on PAD cannot be accessed.
Nevertheless, this study has several strengths. First, the main strength is that few studies have investigated the association between smoking characteristics (including pack-years of smoking and years since quitting smoking) and PAD risk. Our more detailed analysis, compared with previous studies, allowed us to find a significant association between smoking habits and PAD. Second, ABI was measured on both the left and right sides, and the smallest ABI was used in defining PAD. Third, a large number of community-dwelling older men participated in this study. Because subclinical atherosclerosis progresses with age, the present study population (aged 50 years and older) may help detect an association between smoking habits and PAD.