Sudden cardiac death (SCD) is a leading cause of cardiovascular

Smoking, Smoking Cessation, and Risk of Sudden Cardiac Death in Women Roopinder K. Sandhu, MD, MPH; Monik C. Jimenez, ScD; Stephanie E. Chiuve, ScD; Kathryn C. Fitzgerald, MSc; Stacey A. Kenfield, ScD; Usha B. Tedrow, MD, MSc; Christine M. Albert, MD, MPH Background Few prospective studies have examined quantitative cigarette consumption and smoking cessation on sudden cardiac death (SCD) risk with long-term follow-up. Methods and Results We prospectively examined the association between cigarette smoking and smoking cessation on the risk of SCD among 101 018 women participating in the Nurses Health Study without known coronary heart disease, stroke, and cancer at baseline 1980. During 30 years of follow-up, we identified 351 SCD events. Compared with never smokers, current smokers had a 2.44-fold (95% CI, 1.80 3.31) increased risk of SCD after controlling for coronary risk factors. In multivariable analyses, quantity of cigarettes smoked daily (P value for trend, <0.0001) and smoking duration (P value for trend, <0.0001) were linearly associated with SCD risk among current smokers. Small-to-moderate amounts of cigarette consumption (1 14 per day) were associated with a significant 1.84-fold (95% CI, 1.16 2.92) increase in SCD risk and every 5 years of continued smoking was associated with an 8% increase in SCD risk (hazard ratio, 1.08; 95% CI, 1.05 1.12; P<0.0001). The SCD risk linearly decreased over time after quitting and was equivalent to that of a never smoker after 20 years of cessation (P value for trend, <0.0001). Conclusions In this large prospective cohort of women without coronary heart disease at baseline, a strong dose response relationship between cigarette smoking and SCD risk was observed, and smoking cessation significantly reduced and eventually eliminated excess SCD risk. This suggests efforts to prevent SCD among women should include aggressive strategies for smoking cessation. (Circ Arrhythm Electrophysiol. 2012;5:1091-1097.) Key Words: epidemiology risk factors sudden death smoking Sudden cardiac death (SCD) is a leading cause of cardiovascular mortality. 1 Most SCDs occur in those without known cardiac disease and not in the well-recognized highrisk subset with myocardial infarction, impaired left-ventricular ejection fraction, and heart failure. 2,3 Therefore, the current paradigm for SCD prevention does not address the population with the highest number of SCD events. Understanding population-based characteristics that are associated with SCD risk is important for SCD prevention in the general population. Clinical Perspective on p 1097 Cigarette smoking is strongly associated with SCD risk. 4 9 Although smoking status is a well-established risk factor for SCD and coronary heart disease (CHD), 10 12 there are limited data on dose response relationships between quantity of cigarettes smoked, smoking duration, as well as the impact of smoking cessation on SCD risk over a long-term follow-up. Data from prospective studies on dose response relationships appear to conflict. 6,11,13 In the Framingham cohort, 13 quantity of cigarettes smoked daily was linearly associated with SCD risk only among men without underlying CHD, whereas a gradient in risk was not observed among patients with a previous history of CHD in the Bezafibrate Infarction Prevention Trial. 11 These studies were limited by relatively few numbers of SCD events, particularly among women. In addition, among patients with known CHD, smoking cessation has been associated with lower incidence of SCD; 10 12,14 however, sparse data exist about the benefits of smoking cessation among individuals without documented CHD, who comprise the majority of SCD events in the population. 13 To better define the impact of smoking and cessation among women with and without CHD, we prospectively evaluated the association between amount of cigarette smoking (status, quantity of cigarettes smoked daily, and duration of smoking) and smoking cessation on the risk of SCD among 101 018 Received June 14, 2012; accepted November 2, 2012. From the Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, Canada (R.K.S.); Division of Preventive Medicine (R.K.S., M.C.J., S.E.C., U.B.T., C.M.A.), Channing Laboratory (S.A.K.), Cardiovascular Division (U.B.T., C.M.A.), Department of Medicine, Brigham and Women s Hospital, Harvard Medical School, Boston, MA; and Departments of Epidemiology and Nutrition, Harvard School of Public Health, Boston, MA (S.E.C., K.C.F., S.A.K.). The Editor for this article was Kenneth A. Ellenbogen, MD. Correspondence to Roopinder K. Sandhu, MD, Brigham and Women s Hospital, Harvard Medical School, 900 Commonwealth Ave, E Boston, MA, 02215. E-mail rupi.md@gmail.com 2012 American Heart Association, Inc. Circ Arrhythm Electrophysiol is available at http://circep.ahajournals.org 1091 DOI: 10.1161/CIRCEP.112.975219

1092 Circ Arrhythm Electrophysiol December 2012 women free of cardiovascular disease at baseline in the Nurses Health Study over 30 years of follow-up. Methods The Nurses Health Study Cohort The Nurses Health Study was established in 1976, consisting of 121 701 female registered nurses, 30 to 55 years of age who completed a questionnaire with detailed information about medical history, cardiovascular risk factors, menopausal status, and lifestyle factors. 15 Twice a year, follow-up questionnaires are mailed to update exposure information and ascertain newly diagnosed medical conditions. Study Population Follow-up for the present analysis began in 1980 to include important covariates (alcohol use, physical activity, and vitamin use), which were not ascertained until this follow-up period. Women who died before 1980 or with a missing date of death (n=748), had incomplete information about smoking questions (n=4118) or baseline characteristics (n=8634), and previous history of angina, myocardial infarction, coronary revascularization, or stroke or cancer at baseline (n=7183) were excluded leaving 101 018 women who were included in the analysis. Women who reported CHD (angina, myocardial infarction, or coronary revascularization) during follow-up were not excluded from the primary analyses. This study was approved by the Partners Human Research Committee (Boston, MA) and subjects, or family members provided written, informed consent. Assessment of Smoking and Covariates Participants reported whether they currently smoked or had ever smoked in the past and the age at which they started smoking on the initial questionnaire. Current smokers reported the number of cigarettes smoked per day. Past smokers reported the age at which they stopped smoking and the number of cigarettes smoked per day before quitting. At the start of each 2-year follow-up cycle, women were reclassified according to most recent smoking status (never, past, current), quantity of cigarettes smoked per day, duration of smoking, and time since quitting among former smokers. We used cigarettes smoked per day and duration of smoking separately instead of a summary measure of pack-years which can classify women with varying doses and duration as having the same pack-years. 16 Covariates including age (continuous), history of diabetes mellitus (yes versus no), history of hypertension (yes versus no), history of hypercholesterolemia (yes versus no), body mass index (<25, 25 29.9, >30 kg/m 2 ), menopausal status and postmenopausal hormone use (yes versus no), usual aspirin use (<1, 1 5, >6 times/wk), multivitamin use (yes versus no), vitamin E use (yes versus no) were updated at the start of each 2-year follow-up cycle. Information on alcohol intake (0, 0.1 4.9, 5.0 14.9, 15.0 29.9, >30 g/d) and physical activity (0, 0 1.9, 2 3.9, >4 h/wk) was updated every 2 to 4 years and family history of myocardial infarction (no, before age 60 years, after age 60 years) was obtained in 1976 and updated in 1984. End-Point Ascertainment and Definitions The study end points included incident cases of SCD after the return of the 1980 questionnaire and before January 1, 2011. Deaths were reported by either next of kin, postal authorities or identified through a search of the National Death Index and confirmed with death certificates. We confirmed the end point of SCD through review of medical records, autopsy reports, and interviews with family members about the circumstances surrounding the death. SCD classification has been previously described. 6 Briefly, the pathogenesis of cardiac death was considered sudden if the death or cardiac arrest precipitating death occurred within 1 hour of symptoms. Women with documented collapse of circulation (hypotension, exacerbation of congestive heart failure, or altered mental status) before disappearance of pulse were excluded from arrhythmic deaths to increase specificity. 17 Unwitnessed deaths occurring within 1 hour of symptom onset and with autopsy findings consistent with SCD were considered possible SCDs (n=31) and included in the analysis. In a secondary analysis, we included unwitnessed deaths and deaths that occurred during sleep where the women were documented to be asymptomatic in the preceding 24 hours and where circumstances suggested the death could be sudden (n=117). SCD events were confirmed by 2 cardiologists. Statistical Analysis Person-years of follow-up were calculated for each woman from date of return of the 1980 questionnaire to date of death or to January 1, 2011, whichever came first. Women with missing smoking data had nonmissing data carried forward for 1 follow-up cycle. If there was missing data >1 follow-up cycle, person-time was not contributed for respective cycles. All other covariates with missing data had persontime assigned to a missing category for that covariate. In the present analysis, current smokers were classified into categories of cigarettes smoked per day; 1 to 14, 15 to 24, and >25. Current and past smokers were classified as ever smoker to assess duration of smoking which was classified into categories of years: <12, 13 to <24, 25 to <35, and >35. In addition, smoking duration was evaluated as a continuous variable in 5-year increments. For past smokers, cessation was classified into categories of years since quitting of <5, 5 to <10, 10 to <15, 15 to <20, and >20. We used time-varying Cox models to estimate age and multivariable-adjusted hazard ratios (HRs) for SCD across categories of smoking status, cigarettes smoked per day, smoking duration, and years since quitting smoking. Never smoker was the reference group for analyses on smoking status, cigarettes smoked per day, and smoking duration. Current smokers were the reference category for analyses examining smoking cessation among past smokers. Multivariable models adjusted for all covariates. A Wald test was conducted for linear trend by assigning median values to categories of cigarettes smoked per day, duration of smoking (years), and years since quitting and modeling this variable as a continuous variable. CHD was not controlled for in multivariable models because it is a potential intermediate on the causal pathway between smoking and SCD. To evaluate whether the association between smoking and smoking cessation differs once CHD becomes clinically manifest, a prespecified analysis examining effect modification by history of CHD was performed. To test formally for an interaction between CHD and each smoking variable (status, quantity of daily cigarettes, duration, and years since quitting), we added a cross-product term into separate multivariable models and used a likelihood ratio test, comparing a model with and without the interaction term. We then performed separate multivariable models stratified by CHD status before SCD. Statistical analyses were performed by using of SAS software version 9.1.3 (SAS Institute Inc, Cary, NC). All probability values were 2-sided and statistically significant at P<0.05. Results Among the 101 018 women without known CHD, stroke or cancer at baseline, 29.1% were current smokers, 26.4% past smokers, and 44.5% never smokers. Women who reported current smoking were more likely to be postmenopausal and consumed >15.0 g of alcohol per day. They were less likely to have a body mass index >30 kg/m 2, regularly use multivitamin or vitamin E, and participate in physical activity >4 h/wk compared with never smokers (Table 1). During 30 years of follow-up, there were 351 incident SCDs. Current smoking was associated with a 2.44-fold (95% CI, 1.80 3.31) increased risk of SCD and women who quit smoking had a 1.40-fold (95% CI, 1.10 1.79) increased risk of SCD compared with a never smoker in multivariable analysis (Table 2). Compared with women who continued to smoke, women who quit smoking had a lower risk of SCD (HR, 0.58; 95% CI, 0.43 0.77) in multivariable analysis.

Sandhu et al Smoking and Sudden Cardiac Death 1093 Table 1. Age-Standardized Characteristics by Smoking Status Among 101 018 Participants in the Nurses Health Study at Baseline in 1980 Characteristics Never (n=44 904) Past (n=26 684) Current (29 430) P Value Smoking status of participants (%) 44.5 26.4 29.1 Total person-years of follow-up 1 181 691 978 504 396 698 Duration of smoking, mean (SD), y NA 14.7 (9.1) 26.6 (8.1) <0.0001 Age at smoking initiation, mean (SD) NA 15.6 (8.2) 18.1 (6.4) <0.0001 Hypertension (%) 16 17 15 <0.0001 Diabetes mellitus (%) 2 2 2 <0.0001 High cholesterol (%) 5 6 5 <0.0001 BMI>30 kg/m 2 (%) 12 12 8 <0.0001 Parental MI at age <60 y (%) 11 13 13 <0.0001 Menopausal status (%) <0.0001 Premenopausal 61 61 56 Postmenopausal 32 33 38 Current postmenopausal hormone use (%) 7 8 8 <0.0001 Aspirin use 6+ times/wk (%) 23 24 24 <0.0001 Multivitamin use (%) 30 33 28 <0.0001 Vitamin E use (%) 12 13 10 <0.0001 Alcohol intake (%) <0.0001 0 g/d 42 23 25 0.1 4.9 g/d 34 34 32 5.0 14.9 g/d 17 27 26 >15.0 g/d 6 15 18 Physical Activity (h/wk; %) <0.0001 >0 1.9 34 33 37 2.0 3.9 25 23 28 >4 41 45 34 BMI denotes body mass index; and MI, myocardial infarction. In age and multivariable-adjusted models, quantity of cigarettes smoked daily among current smokers was linearly associated with risk of SCD (Table 2; P value for trend <0.0001). Small-to-moderate amounts of daily cigarette consumption (1 14 per day) was significantly associated with a 1.84-fold (95% CI, 1.16 2.92) increase in SCD risk, and women who smoked 25 cigarettes or more had a 3.30- fold (95% CI, 2.04 5.33) increase in SCD risk compared with never smokers in multivariable models. Duration of smoking was also linearly associated with SCD risk (Table 3; P value for trend <0.0001), with the highest risk associated with a smoking duration >35 years in both age and Table 2. Age and Multivariable-Adjusted Relative Risks of SCD According to Smoking Status and Quantity of Cigarettes Smoked Daily for Current Smokers Never (n=44 904) Smoking Status Past (n=26 684) Current (n=29 430) Quantity of Cigarettes Smoked Daily by Current Smokers 1 14 (n=8085) 15 24 (n=12 630) No. of events (n=351) 128 148 75 22 31 22 Person-years of follow-up (n=2 556 893) >25 (n=8715) 1 181 691 978 504 396 698 142 244 161 000 93 454 Incidence rate* 11 15 19 15 19 24 P Value for Trend Age adjusted 1.00 (referent) 1.32 (1.04 1.68) 2.42 (1.81 3.25) 1.69 (1.07 2.66) 2.65 (1.77 3.95) 3.70 (2.32 5.90) <0.0001 Multivariable 1.00 (referent) 1.40 (1.10 1.79) 2.44 (1.80 3.31) 1.84 (1.16 2.92) 2.62 (1.74 3.94) 3.30 (2.04 5.33) <0.0001 SCD indicates sudden cardiac death. *Incidence per 100 000 person-years. Multivariable models adjusted with updated values for age, diabetes mellitus, hypertension, hypercholesterolemia, body mass index, alcohol, physical activity, menopausal status and postmenopausal hormone use, aspirin, multivitamin, vitamin E, and family history of myocardial infarction.

1094 Circ Arrhythm Electrophysiol December 2012 Table 3. Age and Multivariable-Adjusted Relative Risks of SCD According to Years of Smoking Never (n=44 904) <12 (n=1331) 13 <24 (n=20 832) 25 <35 (n=17 363) >35 (n=4588) No. of events (n=351) 128 34 31 47 111 Person-years of follow-up (n=2 556 899) 1 181 565 323 626 366 063 344 573 341 072 Incidence rate* 11 11 8 14 33 P Value for Trend Age adjusted 1.00 (referent) 1.21 (0.83 1.77) 0.94 (0.63 1.39) 1.44 (1.03 2.02) 2.22 (1.71 2.87) <0.0001 Multivariable 1.00 (referent) 1.35 (0.92 1.98) 1.02 (0.69 1.52) 1.48 (1.05 2.08) 2.22 (1.71 2.90) <0.0001 SCD indicates sudden cardiac death. Smoking duration includes ever smoke. *Incidence per 100 000 person-years. Multivariable models adjusted with updated values for age, diabetes mellitus, hypertension, hypercholesterolemia, body mass index, alcohol, physical activity, menopausal status and postmenopausal hormone use, aspirin, multivitamin, vitamin E, and family history of myocardial infarction. multivariable-adjusted models (multivariable HR, 2.22; 95% CI, 1.71 2.90). For every 5 years of continued smoking, there was an associated 8% increase in SCD risk (HR, 1.08; 95% CI, 1.05 1.12; P<0.0001) in multivariable analysis. Compared with current smokers, the HR for SCD linearly decreased among past smokers with length of time since quitting. (Figure; P<0.0001). This risk reduction became significant at >15 years of smoking cessation, and by 20 years, the HR for SCD in comparison with women who continued to smoke was similar for past (HR, 0.45; 95% CI, 0.31 0.64) and never smokers (HR, 0.40; 95% CI, 0.30 0.54). In a secondary analysis, which included unwitnessed deaths where the women were documented to be asymptomatic in the 24 hours preceding death, the association between amount of cigarette smoking and SCD risk was not materially altered. A linear association remained between quantity of cigarettes smoked daily (1 14: HR, 1.74; 95% CI, 1.14 2.66, 15 24: HR, 2.90; 95% CI, 2.04 4.14, >25: HR, 3.12; 95% CI, 1.99 4.90, P value for trend <0.0001) and duration of smoking (<12: HR, 1.35; 95% CI, 0.96 1.88, 13 <24: HR, 1.08; 95% CI, 0.77 1.52, 25 <35: HR, 1.34; 95% CI, 0.97 1.83, >35: HR, 2.42; 95% CI, 1.93 3.03, P value for trend <0.0001) and SCD risk in multivariable analyses. Similarly, SCD risk reduction remained significantly associated with length of time since quitting (P<0.0001). We explored whether the association between smoking status and SCD differed by the presence or absence of established CHD (Table 4). Among women who did not develop CHD, the incidence rate of SCD was 10, 11, and 17 per 100 000 for never, past, and current smokers, respectively. For women who developed CHD, comparable rates were much higher (46, 83, and 78 per 100 000 for never, past, current smokers, respectively). Overall, women with CHD who quit smoking continued to have a higher SCD risk compared with never smokers (HR 1.87, 95% CI, 1.16 3.01); whereas women without CHD did not (P value for interaction=0.15). SCD risk reduction associated with smoking cessation varied significantly among women with and without CHD (P value for interaction=0.03). An immediate reduction in SCD risk at <5 years was observed among women without CHD compared with the risk reduction observed among women with CHD which occurred at 15 to 20 years of smoking cessation. Figure. Age and multivariable-adjusted hazard ratios for SCD according to years since quitting among past smokers. The error bars denote 95% CIs. The reference category is current smokers. The white diamond represents age-adjusted hazard ratio (HR). The black diamond represents multivariable HR adjusted with updated values for age, diabetes mellitus, hypertension, hypercholesterolemia, body mass index, alcohol, physical activity, menopausal status and postmenopausal hormone use, aspirin, multivitamin, vitamin E, and family history of myocardial infarction. P value for trend <0.0001 in the age and multivariable-adjusted models.

Sandhu et al Smoking and Sudden Cardiac Death 1095 Table 4. Multivariable-Adjusted Hazard Ratios of SCD According to Smoking Status, Quantity of Cigarettes, Duration and Smoking Cessation stratified by Clinically Apparent CHD at Time of SCD No. of Cases (n=254) Person-Years of Follow-up (n=2 234 289) No CHD The association between quantity of cigarettes smoked daily and duration of smoking and SCD did not significantly vary among women with and without CHD. Discussion In this large, long-term, prospective cohort of women without known CHD, stroke or cancer at baseline, quantity and duration of smoking were directly associated with SCD. In multivariable analysis, quantity of cigarettes smoked daily was linearly associated with SCD risk with almost a 2.0- fold increase associated with small-to-moderate amounts of cigarette consumption. With respect to smoking cessation, SCD risk reduction linearly increased with length of time since quitting, decreasing to the level of never smoker after 20 years. In stratified analyses, the elevated SCD risk associated with past smoking appeared to be limited to those with established CHD, and the time course of the SCD risk reduction associated with smoking cessation varied significantly among women with and without CHD. An immediate reduction in SCD risk was observed among women without CHD compared with the eventual risk reduction observed among women with CHD. These data about the association between smoking and SCD in women are generally consistent with those reported in other prospective population-based cohort studies. In a British study of 7735 middle-age men, current smoking was associated with SCD, but only in the absence of ischemic heart disease. 9 In the Framingham Study, smoking >20 cigarettes per day was associated with a long-term elevation in SCD risk among women but not among men. 8 Smoking was not associated with IR Multivariable HR (95%) CI No. of Cases (n=97) CHD Person-Years of Follow-up (n=144 553) P Value* an elevation in short-term risk in women or men. Earlier data from this cohort based on only 50 SCD events suggested a lack of a dose response relationship between cigarette smoking and SCD among women. 18 Using a total of 351 SCD events, we were able to demonstrate a strong dose response relationship between quantity of cigarettes smoked daily and SCD risk. Previous studies 5,18 also suggested that a gradient in SCD risk associated with increasing cigarette use was less apparent once CHD becomes clinically manifest, but we were able to demonstrate direct associations between number of cigarettes smoked among women with and without CHD. Our data reinforce the important concept that SCD risk associated with cigarette smoking is modifiable. Smoking cessation has previously been associated with SCD risk reduction primarily in populations with established CHD 10 12,14 ; whereas a paucity of data exists on smoking cessation and SCD among individuals without CHD. In the Framingham study, male smokers who were <65 at study entry and subsequently quit had half the sudden coronary death rate compared with those who continued to smoke over 18 years of follow-up. 19 Similar to the above studies, we observed a reduction in SCD risk associated with smoking cessation in women. In contrast to previous data suggesting a fairly prompt reduction in SCD risk among patients with CHD, 10 12,14 our data suggest that the reduction in SCD risk may be delayed once women develop manifestations of CHD. In this setting, severe and vulnerable coronary atherosclerotic plaques and resultant myocardial fibrosis and scarring may confer a residual SCD risk which persists for some time after smoking cessation. Smoking IR Multivariable HR (95%) CI Smoking status 0.15 Never 100 1 035 670 10 1.00 (Referent) 28 60 371 46 1.00 (Referent) Past 91 828 023 11 1.17 (0.87 1.56) 57 68 833 83 1.87 (1.16 3.01) Current 63 370 596 17 2.43 (1.72 3.42) 12 15 349 78 1.80 (0.84 3.86) No. cigarettes smoked (per category increase) Smoking duration (per every 5 y) 63 370 596 17 1.35 (1.21 1.51) 12 15 349 78 1.33 (1.05 1.67) 0.81 154 1 198 619 28 1.07 (1.03 1.11) 69 84 182 161 1.09 (1.03 1.15) 0.70 Smoking cessation, y 0.03 Current 63 370 596 17 1.00 (Referent) 12 15 349 78 1.00 (Referent) <5 10 114 384 9 0.47 (0.24 0.92) 15 9174 164 2.14 (0.91 5.03) 5 <10 14 108 069 13 0.63 (0.35 1.14) 9 9544 94 1.25 (0.49 3.19) 10 <15 9 107 430 8 0.43 (0.21 0.87) 11 8492 130 1.61 (0.65 3.99) 15 <20 12 116 613 9 0.52 (0.28 0.98) 5 8747 57 0.53 (0.17 1.64) >20 46 381 526 12 0.45 (0.30 0.68) 16 32 877 52 0.59 (0.26 1.35) Never 100 1 035 670 9 0.41 (0.29 0.58) 26 60 371 46 0.51 (0.24 1.10) SCD indicates sudden cardiac death; and CHD, coronary heart disease (angina, myocardial infarction [MI], or coronary revascularization); and IR, incidence rate. *P value for interaction. Incidence per 100 000 person-years. Multivariable models adjusted with updated values for age, diabetes mellitus, hypertension, hypercholesterolemia, body mass index, alcohol, physical activity, menopausal status and postmenopausal hormone use, aspirin, multivitamin, vitamin E, and family history of MI.

1096 Circ Arrhythm Electrophysiol December 2012 cessation behavior after a CHD event may also be a function of the nature of CHD events, which in turn could influence residual SCD risk. Women who quit smoking after a CHD event may have had a more severe event than those who continue to smoke resulting in a continued elevation in SCD risk. In contrast, the more immediate SCD risk reduction observed among women without CHD may, in part, be explained by alleviation of acute proarrhythmic effects of nicotine such as reductions in catecholamine release, 20 alteration of potassium handling, 21 and lowering of ventricular fibrillation threshold caused by transient enhanced platelet adhesion. 20 This study has several strengths, including the prospective design, large number of participants with long follow-up, and outcome events confirmed by next-of-kin reports and medical record review. There are also several limitations of the present study which warrant discussion. First, information on cigarette smoking was based on self-report, which could lead to misclassification. However, numerous validation studies have established the accuracy of self-reported smoking and other CHD risk factors in this cohort of health professionals. 22 24 In addition, smoking characteristics were assessed on 2-year follow-up questionnaires and updated accordingly to further minimize misclassification. Second, determination of SCD is difficult in free-living individuals and is susceptible to error. Accordingly, we tried to minimize error by employing a strict definition of SCD 6 and requiring supporting information from medical records or witnessed reports. These strict criteria would have missed women who died suddenly without supporting information. However, our results were not significantly altered when a sensitivity analysis was performed which included a broader definition of SCD. Third, we did not have direct data about the severity of CHD events to assess the relationship between smoking cessation behavior and CHD events. Fourth, participants of the Nurses Health Study are predominately white (96%), relatively healthy, and of homogeneous socio-demographic status, and it is uncertain whether these results would be generalizable to other ethnicities and socio-demographic groups. Finally, because of the observational nature of the study, we cannot exclude the possibility that residual or unmeasured confounding may have accounted for part of the associations observed. However, the magnitude of observed associations were not altered substantially after controlling for numerous CHD risk factors, and therefore, residual confounding by these traditional CHD risk factors is unlikely. Conversely, there may be unknown, and thus unmeasured, factors which may mediate or confound the association between smoking and SCD. In summary, these prospective data provide further evidence that cigarette smoking is an important modifiable risk factor for SCD among women with and without CHD. A dose-dependent relationship between smoking and SCD was observed, and even small-to-moderate quantities of daily smoking were associated with SCD risk. Because SCD is often the first manifestation of CHD among women, 4,6 lifestyle modifications that reduce SCD in this apparently healthy subset of the population are particularly important. The benefits of smoking cessation on SCD risk in women are in keeping with benefits demonstrated for reductions in all-cause mortality, CHD, and cancer. 25,26 Smoking cessation was associated with significant reductions in SCD risk in women both with and without CHD. The reduction in SCD risk may be more immediate among women without CHD arguing that efforts at smoking cessation should not await the development of CHD. Efforts to prevent SCD among women should consider aggressive strategies for smoking cessation among all women. Sources of Funding This study was supported by research grants CA87969 and HL034594 from the National Institutes of Health and an Established Investigator Award from the American Heart Association to Dr Albert. None. Disclosures References 1. Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, Ferguson TB, Ford E, Furie K, Gillespie C, Go A, Greenlund K, Haase N, Hailpern S, Ho PM, Howard V, Kissela B, Kittner S, Lackland D, Lisabeth L, Marelli A, McDermott MM, Meigs J, Mozaffarian D, Mussolino M, Nichol G, Roger VL, Rosamond W, Sacco R, Sorlie P, Roger VL, Stafford R, Thom T, Wasserthiel-Smoller S, Wong ND, Wylie-Rosett J; WRITING GROUP MEMBERS; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics 2010 update: a report from the American Heart Association. Circulation. 2010;121:e46 e215. 2. Myerburg RJ, Kessler KM, Castellanos A. Sudden cardiac death: epidemiology, transient risk, and intervention assessment. Ann Intern Med. 1993;119:1187 1197. 3. Stecker EC, Vickers C, Waltz J, Socoteanu C, John BT, Mariani R, McAnulty JH, Gunson K, Jui J, Chugh SS. Population-based analysis of sudden cardiac death with and without left ventricular systolic dysfunction: two-year findings from the Oregon Sudden Unexpected Death Study. J Am Coll Cardiol. 2006;47:1161 1166. 4. Kannel WB, Wilson PW, D Agostino RB, Cobb J. Sudden coronary death in women. Am Heart J. 1998;136:205 212. 5. de Vreede-Swagemakers JJ, Gorgels AP, Weijenberg MP, Dubois-Arbouw WI, Golombeck B, van Ree JW, Knottnerus A, Wellens HJ. Risk indicators for out-of-hospital cardiac arrest in patients with coronary artery disease. J Clin Epidemiol. 1999;52:601 607. 6. Albert CM, Chae CU, Grodstein F, Rose LM, Rexrode KM, Ruskin JN, Stampfer MJ, Manson JE. Prospective study of sudden cardiac death among women in the United States. Circulation. 2003;107:2096 2101. 7. Jouven X, Desnos M, Guerot C, Ducimetière P. Predicting sudden death in the population: the Paris Prospective Study I. Circulation. 1999;99: 1978 1983. 8. Cupples LA, Gagnon DR, Kannel WB. Long- and short-term risk of sudden coronary death. Circulation. 1992;85(1 suppl):i11 I18. 9. Wannamethee G, Shaper AG, Macfarlane PW, Walker M. Risk factors for sudden cardiac death in middle-aged British men. Circulation. 1995;91:1749 1756. 10. Peters RW, Brooks MM, Todd L, Liebson PR, Wilhelmsen L. Smoking cessation and arrhythmic death: the CAST experience. The Cardiac Arrhythmia Suppression Trial (CAST) Investigators. J Am Coll Cardiol. 1995;26:1287 1292. 11. Goldenberg I, Jonas M, Tenenbaum A, Boyko V, Matetzky S, Shotan A, Behar S, Reicher-Reiss H; Bezafibrate Infarction Prevention Study Group. Current smoking, smoking cessation, and the risk of sudden cardiac death in patients with coronary artery disease. Arch Intern Med. 2003;163:2301 2305. 12. Vlietstra RE, Kronmal RA, Oberman A, Frye RL, Killip T 3rd. Effect of cigarette smoking on survival of patients with angiographically documented coronary artery disease. Report from the CASS registry. JAMA. 1986;255:1023 1027. 13. Kannel WB, Schatzkin A. Sudden death: lessons from subsets in population studies. J Am Coll Cardiol. 1985;5(6 suppl):141b 149B. 14. Hallstrom AP, Cobb LA, Ray R. Smoking as a risk factor for recurrence of sudden cardiac arrest. N Engl J Med. 1986;314:271 275. 15. Colditz GA, Rimm EB, Giovannucci E, Stampfer MJ, Rosner B, Willett WC. A prospective study of parental history of myocardial infarction and coronary artery disease in men. Am J Cardiol. 1991;67:933 938.

Sandhu et al Smoking and Sudden Cardiac Death 1097 16. Samet JM, Thun MJ, de Gonzalez AB. Models of smoking and lung cancer risk: a means to an end. Epidemiology. 2007;18:649 651. 17. Hinkle LE Jr, Thaler HT. Clinical classification of cardiac deaths. Circulation. 1982;65:457 464. 18. Schatzkin A, Cupples LA, Heeren T, Morelock S, Kannel WB. Sudden death in the Framingham Heart Study. Differences in incidence and risk factors by sex and coronary disease status. Am J Epidemiol. 1984;120:888 899. 19. Gordon T, Kannel WB, McGee D, Dawber TR. Death and coronary attacks in men after giving up cigarette smoking. A report from the Framingham study. Lancet. 1974;2:1345 1348. 20. Mehta MC, Jain AC, Mehta A, Billie M. Cardiac Arrhythmias Following Intravenous Nicotine: Experimental Study in Dogs. J Cardiovasc Pharmacol Ther. 1997;2:291 298. 21. Wang HX, Ng TB. Natural products with hypoglycemic, hypotensive, hypocholesterolemic, antiatherosclerotic and antithrombotic activities. Life Sci. 1999;65:2663 2677. 22. Manson JE, Colditz GA, Stampfer MJ, Willett WC, Krolewski AS, Rosner B, Arky RA, Speizer FE, Hennekens CH. A prospective study of maturityonset diabetes mellitus and risk of coronary heart disease and stroke in women. Arch Intern Med. 1991;151:1141 1147. 23. Fiebach NH, Hebert PR, Stampfer MJ, Colditz GA, Willett WC, Rosner B, Speizer FE, Hennekens CH. A prospective study of high blood pressure and cardiovascular disease in women. Am J Epidemiol. 1989;130:646 654. 24. Willett W, Stampfer MJ, Bain C, Lipnick R, Speizer FE, Rosner B, Cramer D, Hennekens CH. Cigarette smoking, relative weight, and menopause. Am J Epidemiol. 1983;117:651 658. 25. Kawachi I, Colditz GA, Stampfer MJ, Willett WC, Manson JE, Rosner B, Speizer FE, Hennekens CH. Smoking cessation and time course of decreased risks of coronary heart disease in middle-aged women. Arch Intern Med. 1994;154:169 175. 26. Kenfield SA, Stampfer MJ, Rosner BA, Colditz GA. Smoking and smoking cessation in relation to mortality in women. JAMA. 2008;299:2037 2047. CLINICAL PERSPECTIVE Because the majority of sudden cardiac deaths (SCDs) occur in those without known cardiac disease, understanding population-based characteristics that are associated with SCD risk is important for prevention. We sought to examine the prospective association between quantitative cigarette consumption and smoking cessation on the risk of SCD among 101 018 apparently healthy women in the Nurses Health Study over 30 years of follow-up. We also evaluated whether the association between smoking and smoking cessation differs once coronary heart disease (CHD) becomes clinically manifest. Our data suggest a significant dose response relationship between the quantity of cigarettes smoked daily and SCD for women with and without clinical CHD. Even small-to-moderate quantities of daily smoking (1 14 cigarettes per day) were associated with an elevated risk, and the longer women smoked, the higher the SCD risk. Women who quit smoking had a lower SCD risk, which continued to decrease over time and was equivalent to that of a never smoker at 20 years. Women without a history of known CHD had a more immediate reduction in SCD risk after quitting compared with women previously diagnosed with CHD. Therefore, cigarette smoking is an important modifiable risk factor for SCD among women with and without CHD, and cessation should not await the development of CHD. Lifestyle modification to prevent SCD should include aggressive strategies for smoking cessation among all women.