Abstract
Background
Whether smoking increases the risk of atrial fibrillation (AF) remains debatable due to inconsistent reports.
Methods
We examined the association between smoking and incident AF in 11,047 participants from the REasons for Geographic And Racial Differences in Stroke (REGARDS) Study, one of the largest biracial, population-based cohort studies in the USA. Baseline (2003–2007) cigarette smoking status and amount (pack-years) were self-reported. Incident AF was determined by electrocardiography and history of a prior physician diagnosis at a follow-up examination conducted after a median of 10.6 years.
Results
During follow-up, 954 incident AF cases were identified; 9.5% in smokers vs. 7.8% in non-smokers; p<0.001. In a model adjusted for socio-demographics, smoking (ever vs. never) was associated with a 15% increased risk of AF [OR (95%CI):1.15(1.00, 1.31)], but this association was no longer significant after further adjustment for cardiovascular risk factors [OR (95% CI):1.12 (0.97,1.29)]. However, heterogeneities in the association were observed among subgroups; the association was stronger in young vs. old participants [OR (95%CI):1.31 (1.03,1.67) vs. 0.99 (0.83–1.18) respectively; interaction p-value=0.005] and in those with vs. without prior cardiovascular disease [OR (95%CI): 1.18 (0.90, 1.56) vs.1.06 (0.90, 1.25) respectively; interaction p-value 0.0307]. Also, the association was significant in blacks but not in whites [OR (95%CI):1.51 (1.12, 2.05) vs. 0.99 (0.84, 1.16), respectively], but the interaction p-value did not reach statistical significance (interaction p-value=0.65).
Conclusions
The association between smoking and AF is possibly mediated by a higher prevalence of cardiovascular risk factors in smokers, but there is marked heterogeneity in the strength of this association among subgroups which may explain the conflicting results in prior studies.
Keywords: Atrial fibrillation, Smoking, REGARDS Study
Introduction
The association between smoking and atrial fibrillation (AF) remains controversial due to conflicting results from population-based studies. While some studies have shown a strong association [1–3], others did not find a link between smoking and AF [4–8]. Smoking is associated with several key mechanisms in the pathogenesis of AF including cardiac autonomic dysfunction, inflammation, and oxidative stress [9,10]. This makes it plausible that there is an association between smoking and AF. However, small sample sizes, short follow-up, lack of racial diversity, and inadequate ascertainment of AF could explain the conflicting results. To overcome the limitations of previous reports, we examined the association between smoking and incident AF in the REasons for Geographic And Racial Differences in Stroke (REGARDS) study, a large biracial prospective cohort study of men and women with long-term follow up.
Methods
The REGARDS study was designed to identify causes of regional and black-white disparities in stroke mortality [11]. Between January 2003 and October 2007, participants were recruited from the continental USA with oversampling from blacks and the stroke belt (North Carolina, South Carolina, Georgia, Alabama, Mississippi, Tennessee, Arkansas, and Louisiana) using commercially available postal and telephone records. Demographic information and medical histories were obtained using a computer-assisted telephone interview system (CATI). In addition, a brief in-home physical examination was performed approximately 3–4 weeks after the telephone interview. During the in-home visit, blood and urine sample collection, a resting electrocardiogram, medication information, height and weight measurements, and blood pressure recordings were performed. All participants provided written informed consent and the study was approved by the institutional review boards of all participating universities.
Out of the 30,239 participants who were enrolled in the REGARDS baseline visit, 15,521 participants completed a 2nd examination similar to the baseline approximately 10 years after the baseline assessment. Of those who completed the 2nd visit, we excluded participants with baseline AF or missing data on AF, smoking status, or other covariates, leaving a final sample of 11,047 participants.
Age, sex, race, income, and education were determined by self-report at baseline. Baseline fasting blood samples were obtained and assayed for total cholesterol, high-density lipoprotein (HDL) cholesterol, and serum glucose. Body mass index (BMI) was calculated as weight in kilograms divided by the square of the height in meters. Diabetes was defined as a fasting blood glucose level ≥126 mg/dl (or non-fasting blood glucose level ≥200 mg /dl among those failing to fast), a physician diagnosis of diabetes, or the current use of medications for diabetes. Systolic blood pressure measurements ≥140 mmHg and/or ≥90 mmHg diastolic blood pressure, or self-reported antihypertensive medication use defined hypertension. History of cardiovascular disease (CVD) was defined as the composite of coronary heart disease, stroke, and peripheral vascular disease. Left ventricular hypertrophy was defined from the electrocardiogram using Sokolow-Lyon criteria.
Baseline cigarette smoking status and amount (pack-years) were self-reported. Participants were asked whether they ever smoked cigarettes and based on their responses, they were categorized as current, former, or never smokers. Additionally, cigarette-years (pack-year) of smoking were calculated for smokers.
Incident AF was identified by the study electrocardiogram and self-reported history of a previous physician diagnosis during the CATI survey at the 2nd in-home visit among those who did not have AF at baseline using the same methods [12]. The electrocardiograms were read and coded at a central reading center (Epidemiological Cardiology Research Center, Wake Forest School of Medicine, Winston Salem, NC, USA) by analysts who were masked to other REGARDS data. Self-reported AF was defined as an affirmative response to the following question: “Has a physician or health professional ever told you that you had atrial fibrillation?”
Baseline characteristics were compared by smoking status (ever vs. never). Categorical variables were reported as frequency and percentage while continuous variables were reported as mean ± standard deviation (SD). Statistical significance for categorical variables was based on the chi-square test of association and the student’s t-test procedure for continuous variables. Multivariable logistic regression was used to compute odds ratios (ORs) and 95% confidence interval (CIs) for association between smoking (ever vs. never; and for current and former vs. never, separately) and incident AF. In smokers, additional models examining the association between pack-years (per 10 pack-year increase) and AF were also fitted. Models were analyzed as follows: Model 1: adjusted for age, sex, race, education, income, and geographic region; Model 2: included covariates in Model 2 with the addition of systolic blood pressure, HDL cholesterol, total cholesterol, body mass index, diabetes, antihypertensive medications, left ventricular hypertrophy, prior CVD, and statin use. To test for the consistency of the results, we conducted subgroup analyses stratified by median age (63 years), sex, race (black vs. white), median level of systolic blood pressure (123 mmHg), diabetes, and history of CVD. Tests for interaction were conducted using Model 2. Statistical significance of all comparisons including interactions was defined as p<0.05. Statistical analyses were performed using SAS version 9.4 (SAS, Cary, NC, USA).
Results
This analysis included 11,047 participants (mean age: 62.3 years; 55% women; 35% blacks). Table 1 shows the baseline characteristics of the participants stratified by smoking status. Smokers (current and former) were more likely to be older, male, with low educational attainment and income, and with a higher prevalence of cardiovascular risk factors such as diabetes, left ventricular hypertrophy, prior CVD, and higher levels of blood lipids and blood pressure, than participants who never smoked.
Table 1.
Baseline characteristics of the study participants
| Characteristic mean± standard deviation or n(%) |
Smoking Status
|
p-value | |
|---|---|---|---|
| Never Smoker (n=5,570) | Ever Smoker (n=5,477) | ||
| Atrial Fibrillation | 432 (7. %) | 522 (9.5%) | <0.001 |
| Age (Years) | 62.82±8.80 | 63.19±7.90 | <0.001 |
| Women | 3530 (63.4%) | 2556 (46.8%) | <0.001 |
| Black | 1980 (35.6%) | 1952 (35.6%) | 0.92 |
| Education | <0.001 | ||
| ≥ College | 2642 (47.4%) | 2174 (36.7%) | |
| Some College | 1328 (23.8%) | 1558 (28.5%) | |
| High School | 1250 (22.4%) | 1320 (24.1%) | |
| Less than HS | 350 (6.3%) | 425 (7.8%) | |
| Income | <0.001 | ||
| $>75,000 | 1291 (23.2%) | 1130 (9.65%) | |
| $35,000–75,000 | 1845 (33.1%) | 1948 (35.6%) | |
| $20,000–$35,000 | 1170 (21.0%) | 1223 (22.3%) | |
| <$20,000 | 632 (11.4%) | 653 (11.9%) | |
| Geographic Region | 0.036 | ||
| Stroke Belt | 1906 (34.2%) | 1796 (32.8%) | |
| Stroke Buckle | 1253 (22.5%) | 1177 (21.5%) | |
| Non-Belt | 2411 (43.3%) | 2504 (45.7%) | |
| Body mass index (kg/m2) | 29.20±6.11 | 29.12±5.56 | <0.001 |
| Hypertension | 2759 (49.5%) | 2768 (50.5%) | 0.29 |
| Systolic blood pressure (mmHg) | 124.69±15.30 | 125.91±15.52 | <0.001 |
| Diastolic blood pressure(mmHg) | 76.05±9.10 | 76.34±9.20 | <0.001 |
| Use of blood pressure medication | 2609 (46.8%) | 2613 (47.7%) | 0.36 |
| Diabetes | 839 (15.1%) | 913 (16.7%) | 0.021 |
| Total cholesterol (mg/dL) | 193.99±38.23 | 191.35±38.78 | <0.001 |
| HDL cholesterol(mg/dL) | 53.84±15.95 | 51.16±15.90 | <0.001 |
| Statin use | 1503 (27.0%) | 1824 (33.3%) | <0.001 |
| Left ventricular hypertrophy | 495 (8.9%) | 402 (7.3%) | 0.003 |
| Prior Cardiovascular disease | 822 (14.8%) | 1079 (19.7%) | <0.001 |
Diabetes was defined as a fasting glucose ≥126 mg/dL (or a non-fasting glucose ≥200 mg/dL among those failing to fast) or self-reported diabetes medication use. Hypertension was defined as systolic blood pressure ≥140 mmHg or a diastolic blood pressure ≥90 mmHg, or by the self-reported use of antihypertensive medications; Stroke belt (North Carolina, South Carolina, Georgia, Alabama, Mississippi, Tennessee, Arkansas, and Louisiana); Stroke Buckle (Coastal plain regions of North Carolina, South Carolina, Georgia); and non-belt is the rest of the United States. HDL= high density lipoprotein
Over a mean follow up of 10.6 years, 954 incident AF cases were identified (9.5% in smokers vs. 7.8% in non-smokers; p<0.001). In a socio-demographic model, smoking (ever vs. never) was associated with a 15% increased risk of AF (p-value = 0.018). However, after further adjustment for cardiovascular risk factors, the association was not significant [OR (95%CI): 1.12 (0.97–1.29)] (Table 2). Similar direction of the results was observed when smoking status was entered in the model as current, former, and never smoker (Supplemental Table 1).
Table 2.
Association between smoking status and AF
| Smoking Status | Participants (n) | AF (n) | Odds Ratio (95% Confidence interval)
|
|
|---|---|---|---|---|
| Model 1* | Model 2† | |||
| Never smoker | 5570 | 432 | 1.0 (Ref) | 1.0 (Ref) |
| Smoker (former and current) | 5477 | 522 | 1.15 (1.00–1.31) | 1.12 (0.97–1.29) |
AF, atrial fibrillation.
Model 1 adjusted for age, sex, race, education, income, geographic region.
Model 2 adjusted for covariates in Model 1 plus systolic blood pressure, high-density lipoprotein cholesterol, total cholesterol, body mass index, diabetes, antihypertensive medications, left ventricular hypertrophy, prior cardiovascular disease, and statin use.
Current smokers were more likely to smoke more (smoking 30 pack-years on average) compared with former smokers (smoking 19 pack-years years on average) (p<0.001). Among smokers, increasing pack-years of smoking was associated with an increased risk of AF in the sociodemographic models, but was no longer significant after adjusting for cardiovascular risk factors (Table 3).
Table 3.
Association between number of pack-years smoked and incident atrial fibrillation among smokers
| Smoking Status | Participants (n) | AF(n) | Odds Ratio (95% Confidence Interval) per 10 pack-year increase
|
|
|---|---|---|---|---|
| Model 1* | Model 2†† | |||
| Current | 1193 | 91 | 1.02 (0.93–1.12) | 0.99 (0.90–1.08) |
| Former | 4284 | 431 | 1.06 (1.02–1.11) | 1.03 (0.99–1.08) |
| All Smokers | 5477 | 522 | 1.05 (1.01–1.09) | 1.03 (0.99–1.07) |
AF, atrial fibrillation.
Model 1 adjusted for age, sex, race, education, income, geographic region.
Model 2 adjusted for covariates in Model 1 plus systolic blood pressure, high-density lipoprotein cholesterol, total cholesterol, body mass index, diabetes, antihypertensive medications, left ventricular hypertrophy, prior cardiovascular disease, and statin use.
In subgroup analyses, heterogeneities in the association between smoking (ever vs. never) and AF were observed. The association was stronger in young vs. old participants (OR (95%CI): 1.31 (1.03, 1.67) vs. 0.99 (0.83–1.18) respectively; interaction p-value=0.005), and in those with vs. without prior CVD [OR (95%CI): 1.18 (0.90, 1.56) vs. 1.06 (0.90, 1.25) respectively; interaction p-value = 0.031]. Also, the association was significant in blacks but not whites [OR (95%CI): 1.51 (1.12, 2.05) vs. 0.99 (0.84, 1.16), respectively)], but the interaction p-value did not reach significance. No differences were observed by sex, median systolic blood pressure, or diabetes subgroups (Fig. 1). Similar results were observed when smoking status was entered in the models as current, former, and never (Supplemental Table 2).
Figure 1.
Smoking and risk of atrial fibrillation in subgroups.
Model adjusted for age, sex, race, education, income, geographic region, systolic blood pressure, high-density lipoprotein cholesterol, total cholesterol, body mass index, diabetes, antihypertensive medications, left ventricular hypertrophy, and prior cardiovascular disease, and statin use.
Discussion
In this analysis from the REGARDS study, one of the largest US biracial cohort studies, we showed that the association between smoking and AF is largely influenced by CVD risk factors, and that there is marked heterogeneity in this association between subgroups.
Attenuation of the association between smoking and AF after adjustment for cardiovascular risk factors suggests that the association between smoking and AF may be mediated by these risk factors in smokers. Smoking is known to enhance atherosclerosis and potentiate the impact of CVD risk factors [13]. Therefore, the impact of smoking on risk factors could be one of the mechanisms that lead to AF in smokers, and hence adjusting for these mediating factors would expectedly attenuate the association. This does not preclude the importance of other mechanisms involved in the pathogenesis of AF development. For example, sympathomimetic effects of nicotine result in increased heart rate and blood pressure by release of catecholamines [14]. Additionally, nicotine has been shown to block transient outward K+ channels [15], and blockage of these channels increases instability by altering membrane transport and cardiac repolarization. Furthermore, nicotine results in the downregulation of microRNA and upregulation of transforming growth factors that leads to proarrhythmic atrial fibrosis [16].
We also observed marked heterogeneity in the association between smoking and AF between age and history of CVD subgroups. This may explain the conflicting results in prior studies. For example, in the Atherosclerosis Risk in Communities (ARIC) Study, which included biracial participants younger than 64 years, smoking was significantly associated with AF [1], similar to the subgroup younger than 63 years in our study. In contrast, smoking was not associated with AF in the Cardiovascular Health Study (participants’ age >65 years) [4], again similar to the older group in our study. Also, we observed a significant association between smoking and AF in participants with prior CVD, and this was found in other studies of participants with prior CVD [1–3].
The differential association between smoking and AF by subgroups, and variation in study participants in prior studies likely explain the conflicting results among studies. Accordingly, these factors must be considered in future studies with aims to examine the association between smoking and AF. Despite these inconsistent findings, conventional wisdom and collective evidence support an association between smoking and AF. Therefore providers should educate patients on the harmful effects of smoking to influence the occurrence of cardiac rhythm disorders. Additionally, although it is unknown if this harm is direct or mediated through other factors, our data support that the relationship between smoking and AF is more pronounced in certain subgroups.
The strengths of this study include the use of a large, biracial cohort with well-ascertained study variables and more than 10 years of follow-up. Additionally, the geographic and racial diversity of the REGARDS cohort provides excellent generalizability of the results. There also are several limitations that deserve consideration. Since some cases of AF may remain undetectable if they were brief (paroxysmal) or not severe, we possibly missed some AF cases. Also, we relied on self-reported questionnaires to obtain some of the baseline characteristics which subjected our study to recall bias. Although we adjusted for several risk factors, residual confounding remains a possibility. Furthermore, our analyses possibly lacked statistical power to detect differences between certain subgroups.
Conclusions
In this analysis from the REGARDS study we showed that smoking is associated with AF after adjusting for sociodemographic characteristics of the individuals, but this association became non-significant after adjusting for cardiovascular risk factors suggesting that it is possibly mediated by a higher prevalence of cardiovascular risk factors in smokers. We also showed that there is marked heterogeneity in the strength of the association between smoking and AF among subgroups which may explain the conflicting results in previous studies.
Supplementary Material
Highlights.
Smoking is associated with atrial fibrillation (AF) but this association became non-significant after adjusting for cardiovascular risk factors suggesting that it is possibly mediated by a higher prevalence of cardiovascular risk factors in smokers.
There is marked heterogeneity in the strength of this association among subgroups which may explain the conflicting results in prior studies.
Acknowledgments
Funding
This research project is supported by a cooperative agreement U01NS041588 from National Institute of Neurological Disorders and Stroke, National Institutes of Health, Department of Health and Human Service. WTO is supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under award F32HL134290. The content is solely the responsibility of the authors and does not necessarily represent official views of National Institute of Neurological Disorders and Stroke or National Institute of Health. The authors thank the other investigators, the staff, and participants of the REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at http://www.regardsstudy.org.
Footnotes
Disclosures: No conflict of interest
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Chamberlain A, Agarwal SK, Folsom AR, Duval S, Soliman EZ, Ambrose M, Eberly LE, Alonso A. Smoking and incidence of atrial fibrillation: results from the Atherosclerosis Risk in Communities (ARIC) study. Heart Rhythm. 2011;8:1160–6. doi: 10.1016/j.hrthm.2011.03.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Heeringa J, Kors JA, Hofman A, van Rooij FJA, Witteman JCM. Cigarette smoking and risk of atrial fibrillation: The Rotterdam Study. Am Heart J. 2008;156:1163–9. doi: 10.1016/j.ahj.2008.08.003. [DOI] [PubMed] [Google Scholar]
- 3.Suzuki S, Otsuka T, Sagara K, Kano Y, Matsuno S, Takai H, Kato Y, Uejima T, Oikawa Y, Nagashima K, Kirigaya H, Kunihara T, Yajima J, Sawada H, Aizawa T, et al. Association between smoking habits and the first-time appearance of atrial fibrillation in Japanese patients: Evidence from the Shinken Database. J Cardiol. 2015;66:73–9. doi: 10.1016/j.jjcc.2014.09.010. [DOI] [PubMed] [Google Scholar]
- 4.Psaty BM, Manolio TA, Kuller LH, White R, Furberg CD, Rautaharju PM. Incidence of and risk factors for atrial fibrillation in older adults. Circulation. 1997;96:2455–61. doi: 10.1161/01.cir.96.7.2455. [DOI] [PubMed] [Google Scholar]
- 5.Stewart S, Hart CL, Hole DJ, McMurray JJV. Population prevalence, incidence, and predictors of atrial fibrillation in the Renfrew/Paisley study. Heart. 2001;86:516–21. doi: 10.1136/heart.86.5.516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med. 1995;98:476–84. doi: 10.1016/S0002-9343(99)80348-9. [DOI] [PubMed] [Google Scholar]
- 7.Wilhelmsen L, Rosengren A, Lappas G. Hospitalizations for atrial fibrillation in the general male population: morbidity and risk factors. J Intern Med. 2001;250:382–9. doi: 10.1046/j.1365-2796.2001.00902.x. [DOI] [PubMed] [Google Scholar]
- 8.Knuiman M, Briffa T, Divitini M, Chew D, Eikelboom J, McQuillan B, Hung J. A cohort study examination of established and emerging risk factors for atrial fibrillation: the Busselton Health Study. Eur J Epidemiol. 2014;29:181–90. doi: 10.1007/s10654-013-9875-y. [DOI] [PubMed] [Google Scholar]
- 9.Felber Dietrich D, Schwartz J, Schindler C, Gaspoz JM, Barthélémy JC, Tschopp JM, Roche F, von Eckardstein A, Brändli O, Leuenberger P, Gold DR, Ackermann-Liebrich U SAPALDIA-team. Effects of passive smoking on heart rate variability, heart rate and blood pressure: an observational study. Int J Epidemiol. 2007;36:834–40. doi: 10.1093/ije/dym031. [DOI] [PubMed] [Google Scholar]
- 10.Yanbaeva DG, Dentener MA, Creutzberg EC, Wesseling G, Wouters EFM. Systemic effects of smoking. Chest. 2007;131:1557–66. doi: 10.1378/chest.06-2179. [DOI] [PubMed] [Google Scholar]
- 11.Howard VJ, Cushman M, Pulley L, Gomez CR, Go RC, Prineas RJ, Graham A, Moy CS, Howard G. The reasons for geographic and racial differences in stroke study: Objectives and design. Neuroepidemiology. 2005;25:135–43. doi: 10.1159/000086678. [DOI] [PubMed] [Google Scholar]
- 12.Soliman EZ, Howard G, Meschia JF, Cushman M, Muntner P, Pullicino PM, McClure LA, Judd S, Howard VJ. Self-reported atrial fibrillation and risk of stroke in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study. Stroke. 2011;42:2950–3. doi: 10.1161/STROKEAHA.111.621367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Benowitz NL. Cigarette smoking and cardiovascular disease: pathophysiology and implications for treatment. Prog Cardiovasc Dis. 2003;46:91–111. doi: 10.1016/s0033-0620(03)00087-2. [DOI] [PubMed] [Google Scholar]
- 14.Haass M, Kubler W. Nicotine and sympathetic neurotransmission. Cardiovasc Drugs Ther. 1997;10:657–65. doi: 10.1007/BF00053022. [DOI] [PubMed] [Google Scholar]
- 15.Wang HZ, Shi H, Zhang LM, Pourrier M, Yang B, Nattel S, Wang Z. Nicotine is a potent blocker of the cardiac A-type K(+) channels. Effects on cloned Kv4.3 channels and native transient outward current. Circulation. 2000;102:1165–71. doi: 10.1161/01.cir.102.10.1165. [DOI] [PubMed] [Google Scholar]
- 16.Shan HL, Zhang Y, Lu YJ, Zhang Y, Pan Z, Cai B, Wang N, Li X, Feng T, Hong Y, Yang B. Downregulation of miR-133 and miR-590 contributes to nicotine-induced atrial remodelling in canines. Cardiovasc Res. 2009;83:465–72. doi: 10.1093/cvr/cvp130. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.