Skip to main content
PMC home page
European Heart Journal logoLink to European Heart Journal
. 2020 Apr 16;41(35):3304–3310. doi: 10.1093/eurheartj/ehaa193

Genetic predisposition to smoking in relation to 14 cardiovascular diseases

Susanna C Larsson 1,2,✉, Amy M Mason 3, Magnus Bäck 4,5, Derek Klarin 6,7,8,9, Scott M Damrauer 10,11; Million Veteran Program, Karl Michaëlsson 12, Stephen Burgess 13,14
PMCID: PMC7544540  PMID: 32300774

Abstract

Aims

The aim of this study was to use Mendelian randomization (MR) to determine the causality of the association between smoking and 14 different cardiovascular diseases (CVDs).

Methods and results

Our primary genetic instrument comprised 361 single-nucleotide polymorphisms (SNPs) associated with smoking initiation (ever smoked regularly) at genome-wide significance. Data on the associations between the SNPs and 14 CVDs were obtained from the UK Biobank study (N = 367 643 individuals), CARDIoGRAMplusC4D consortium (N = 184 305 individuals), Atrial Fibrillation Consortium (2017 dataset; N = 154 432 individuals), and Million Veteran Program (MVP; N = 190 266 individuals). The main analyses were conducted using the random-effects inverse-variance weighted method and complemented with multivariable MR analyses and the weighted median and MR-Egger approaches. Genetic predisposition to smoking initiation was most strongly and consistently associated with higher odds of coronary artery disease, heart failure, abdominal aortic aneurysm, ischaemic stroke, transient ischaemic attack, peripheral arterial disease, and arterial hypertension. Genetic predisposition to smoking initiation was additionally associated with higher odds of deep vein thrombosis and pulmonary embolism in the UK Biobank but not with venous thromboembolism in the MVP. There was limited evidence of causal associations of smoking initiation with atrial fibrillation, aortic valve stenosis, thoracic aortic aneurysm, and intracerebral and subarachnoid haemorrhage.

Conclusion

This MR study supports a causal association between smoking and a broad range of CVDs, in particular, coronary artery disease, heart failure, abdominal aortic aneurysm, ischaemic stroke, transient ischaemic attack, peripheral arterial disease, and arterial hypertension.

graphic file with name ehaa193f4.jpg

Open in a new tab

Keywords: Cardiovascular disease, Lifestyle, Mendelian randomization, Risk factors, Single-nucleotide polymorphisms, Smoking

See page 3311 for the editorial comment on this article (doi: 10.1093/eurheartj/ehaa285)

Introduction

A large body of evidence from prospective observational studies indicates that smoking is a risk factor for several cardiovascular diseases (CVDs),1–6 but the role of smoking for certain CVDs, such as aortic valve stenosis,7–9 atrial fibrillation,10 thoracic aortic aneurysm,11  ,  12 and haemorrhagic stroke3 are limited or inconsistent. In contrast to prospective cohort studies,1 short-term and long-term follow-up of randomized smoking cessation studies have not revealed a significant effect on fatal coronary heart disease events.13  ,  14 Given that much of available data on smoking and CVD derive from observational studies, which are unable to fully account for confounding and reverse causality, and the lack of significant effect of smoking cessation on cardiovascular events in interventional trials, the causal nature of the association between smoking and different CVDs remains to be established.

Mendelian randomization (MR) is an analytical method that uses genetic variants, generally single-nucleotide polymorphisms (SNPs), as unbiased proxy indicators for the modifiable risk factor to determine whether the risk factor is a cause of the disease.15  ,  16 Given that allelic variants are randomly allocated and fixed at conception, MR studies evade reverse causality and are less susceptible to confounding compared with conventional observational studies. To the best of our knowledge, the causal association between smoking and a broad range of CVDs has not been established using MR. We, therefore, applied the MR design to determine the association between smoking and 14 CVD outcomes.

Methods

Study design

Summary-level data (i.e. beta coefficients and standard errors) of the associations between SNPs strongly associated with smoking initiation (primary exposure) and lifetime smoking (secondary exposure) were extracted from the hitherto largest genome-wide association studies for these phenotypes (Figure 1). The corresponding data for the associations of the smoking-associated SNPs with 14 CVDs were available from UK Biobank (Figure 1). In complementary analyses, we obtained summary-level data from genetic consortia of coronary artery disease17 atrial fibrillation,18 and venous thromboembolism.19 Details of the selection of instrumental variables and data sources are provided below.

Figure 1.

Figure 1

Open in a new tab

Summary of data sources and methods used in this study. AF, atrial fibrillation; AFGen, Atrial Fibrillation Consortium; CAD, coronary artery disease; CVD, cardiovascular disease; IVW, inverse-variance weighted; MR, Mendelian randomization; MVP, Million Veteran Program; SNP, single-nucleotide polymorphism; VTE, venous thromboembolism. aData extracted were beta coefficients with corresponding standard errors of the SNP–smoking and SNP–CVD associations.

Genetic instruments

Instrumental variables for smoking initiation were acquired from a genome-wide association meta-analysis of 29 studies, comprising a total of 1 232 091 European-descent individuals of which 383 631 individuals were part of the UK Biobank. That meta-analysis identified 378 conditionally independent genome-wide significant SNPs at 259 loci (defined as a 1MB region nearby the top P-value) associated with smoking initiation, a binary phenotype representing whether an individual had ever smoked cigarettes regularly in their life (current or former).20 Information about pipes/cigar/chew or other non-cigarette forms of tobacco use was not included.20 The smoking-associated SNPs explained 2.3% of the phenotypic variation.20 All but one of the SNPs were available in the UK Biobank. Linkage disequilibrium (defined as r  2 > 0.1 in European populations) between SNPs was evaluated using LDlink21 and was detected among 16 SNP pairs. The SNP with the largest P-value was omitted, leaving 361 SNPs as instrumental variables for smoking initiation. Genetic associations with smoking were reported in standard deviation units.20 Hence, the reported odds ratios correspond to the increase of 1 SD in prevalence of smoking initiation.

In a supplementary analysis, a genetic instrument for lifetime smoking exposure that takes into account smoking status as well as smoking duration, heaviness, and cessation in ever smokers was applied.22 This genetic instrument consists of 126 genome-wide significant SNPs associated with lifetime smoking exposure in UK Biobank (N = 462 690).22

Outcome data

Summary-level data for the smoking-associated SNPs with the 14 CVDs came from the UK Biobank, a cohort study of about 500 000 adults, aged 37–73 years and enrolled between 2006 and 2010.23 We excluded non-White European participants (to minimize confounding by ancestry), those with relatedness of third degree or higher, low genotype call rate (three or more standard deviations from the mean) and excess heterozygosity, resulting in a final study sample of 367 643 individuals. In this study, participants were followed up until 31 March 2017 or the date of death (recorded until 14 February 2018). The median follow-up was 8.0 years. Outcomes were defined based on electronic health records, hospital procedure codes, and self-reported information validated by interview with a nurse (Supplementary material online, Table S1). Logistic regression with adjustment for ten genetic principal components was applied to obtain the beta coefficients and standard errors for the SNP–CVD associations. The SNPs exploited as instrumental variables and their associations with each CVD outcome are presented in Supplementary material online, Table S2.

Summary-level data for coronary artery disease, atrial fibrillation, and venous thromboembolism were additionally obtained from the Coronary ARtery DIsease Genome-wide Replication and Meta-analysis plus The Coronary Artery Disease Genetics (CARDIoGRAMplusC4D) consortium (60 801 cases and 123 504 non-cases),17 the Atrial Fibrillation Consortium (22 346 cases and 132 086 non-cases; AFGen exome chip analysis, not including UK Biobank),18 and the Million Veteran Program (MVP) (8929 cases and 181 337 non-cases), respectively.19 There was no participant overlap between those three datasets and the UK Biobank. For comparison, we also present previously reported results for ischaemic stroke based on the MEGASTROKE consortium (34 217 ischaemic stroke cases and 404 630 non-cases, not including UK Biobank).24  ,  25

The UK Biobank study was approved by the North West Multicenter Research Ethics Committee. Original studies included in the consortia had been approved by a relevant review board. The present analyses were approved by the Swedish Ethical Review Authority.

Statistical analysis

The inverse-variance-weighted (IVW) method, under a multiplicative random-effects model, was used to obtain the primary causal estimates. Ratio estimates are calculated for each SNP as the beta coefficient for the SNP–CVD association divided by the beta coefficient for the SNP–smoking association. These estimates are then combined across SNPs in a random-effects IVW meta-analysis. This MR method provides the highest precision but assumes that all SNPs are valid instrumental variables. Heterogeneity among estimates based on individual SNPs was assessed with the I  2 statistic.26

Multivariable MR analyses27 were performed to adjust for potential confounders that are genetically correlated with smoking initiation, including alcohol drinking (r  g = 0.36), body mass index (r  g = 0.12), and years of education (r  g = −0.40).20 Multivariable MR analysis was also used to assess whether the associations between genetic predisposition to smoking initiation and CVD outcomes that are comorbid or secondary to coronary artery disease28 may be driven by coronary artery disease.

The weighted median and MR-Egger regression methods were used as sensitivity analyses. The weighted median approach gives consistent estimates if at least 50% of the weight in the analysis comes from valid instrumental variables.29 The MR-Egger approach can detect and correct for directional pleiotropy but suffers from low power.29 In a further sensitivity analysis, the associations between genetic predisposition to smoking initiation and CVD outcomes were assessed in never smokers (based on self-report) in UK Biobank.

To correct for multiple testing, Bonferroni correction was applied and two-sided P-values <3.6× 10−3 (where α = 0.05/14 CVD outcomes) were deemed statistically significant and also considered strong evidence of a causal association. Findings with P-values between 0.05 and 3.6 × 10−3 were regarded as suggestive evidence of association. The analyses were performed using the mrrobust30 and MendelianRandomization31 packages.

Results

Participants

In the analytic sample of 367 643 UK Biobank participants, the mean age was 57 years (5th–95th percentile: 43–69 years) and 46% were men. The prevalence of smoking initiation (ever smokers) was 46%. Around 93% were alcohol drinkers.

Smoking and cardiovascular disease

In the univariable IVW analysis, genetic predisposition to smoking initiation was statistically significantly positively associated with 10 of the 14 outcomes in UK Biobank (Figure 2). The odds ratios ranged from 1.18 (95% confidence interval 1.09–1.28; P = 7.1 × 10−5) for atrial fibrillation to 1.81 (95% confidence interval 1.55–2.11; P = 5.2 × 10−7) for peripheral arterial disease. In the multivariable IVW models adjusting for genetically correlated phenotypes (i.e. alcohol consumption, body mass index, and education), the associations remained statistically significant for coronary artery disease, heart failure, abdominal aortic aneurysm, deep vein thrombosis, pulmonary embolism, peripheral arterial disease, and arterial hypertension; suggestive evidence of associations remained for ischaemic stroke, transient ischaemic attack, and atrial fibrillation (Supplementary material online, Table S3). The association of genetic predisposition to smoking initiation with atrial fibrillation did not persist after adjustment for coronary artery disease (Supplementary material online, Table S3). The results were similar in the weighted median analysis and no evidence of directional pleiotropy was detected in the MR-Egger regression analysis (all P ≥ 0.15) (Supplementary material online, Table S3). In the analysis confined to never smokers, all associations except the association with arterial hypertension were null in the univariable or multivariable IVW models (Supplementary material online, Table S4). Genetic predisposition to smoking initiation was also positively associated with coronary artery disease and ischaemic stroke in the CARDIoGRAMplusC4D consortium and MEGASTROKE,25 respectively, but the associations were weaker than in UK Biobank (Supplementary material online, Table S3). There were no robust associations of genetic predisposition to smoking initiation with atrial fibrillation or venous thromboembolism in the AFGen and MVP, respectively (Supplementary material online, Table S3). Findings for the lifetime smoking exposure in UK Biobank (Figure 3) and the consortia and MVP (Supplementary material online, Table S5) were broadly similar to those for smoking initiation.

Figure 2.

Figure 2

Open in a new tab

Associations of genetic predisposition to smoking initiation with 14 cardiovascular diseases in UK Biobank. The odds ratios correspond to the increase of one standard deviation in prevalence of smoking initiation (ever smoked regularly). Estimates are from the multiplicative random-effects inverse variance-weighted method. CI, confidence interval; OR, odds ratio. The I  2 statistic quantifies the amount of heterogeneity among estimates based on individual SNPs. a Significant at the Bonferroni-corrected threshold of P < 3.6 × 10−3.

Figure 3.

Figure 3

Open in a new tab

Associations of genetically predicted lifetime smoking index with 14 cardiovascular diseases in UK Biobank. Odds ratios are expressed per one standard deviation increase of the lifetime smoking index. Estimates are from the multiplicative random-effects inverse variance-weighted method. CI, confidence interval; OR, odds ratio. The I  2 statistic quantifies the amount of heterogeneity among estimates based on individual SNPs individual SNPs. aSignificant at the Bonferroni-corrected threshold of P < 3.6 × 10−3.

For the observed smoking initiation–CVD associations in UK Biobank (Take home figure), we estimated the predicted risk reduction that would result from a reduction in the prevalence of smoking from 46% (the observed prevalence in UK Biobank) to 20%. This corresponds to a 1.22 unit change in the log-odds of smoking, a difference similar to the difference in log odds between those in the 1st and 99th percentiles of the genetic risk score for smoking initiation. This risk reduction was 52% for peripheral arterial disease, 49% for abdominal aortic aneurysm, 41% for heart failure, 32% for coronary artery disease, 29% for pulmonary embolism, 28% for transient ischaemic attack, 27% for ischaemic stroke and deep vein thrombosis, and 21% for arterial hypertension.

Take home figure.

Take home figure

Open in a new tab

 Observed associations of genetic predisposition to smoking initiation and lifetime smoking with cardiovascular diseases in UK Biobank. aNo robust association was observed between smoking and venous thromboembolism (pulmonary embolism and deep vein thrombosis combined) in the Million Veteran Program.

Discussion

Principal findings

This MR study showed that genetic predisposition to smoking is associated with an increased risk of a broad range of CVDs (Take home figure). The strongest and most consistent positive associations were observed between smoking initiation and coronary artery disease, heart failure, abdominal aortic aneurysm, ischaemic stroke, transient ischaemic attack, peripheral arterial disease, and arterial hypertension. Smoking initiation was additionally associated with deep vein thrombosis and pulmonary embolism in the UK Biobank but not with venous thromboembolism in the MVP. There was limited evidence of causal associations of smoking initiation with atrial fibrillation, aortic valve stenosis, thoracic aortic aneurysm, and intracerebral and subarachnoid haemorrhage.

The results from the present MR study based on data from UK Biobank corroborate the findings of conventional prospective observational studies showing that smoking is a risk factor for coronary artery disease,1 heart failure,4 abdominal aortic aneurysm,5 ischaemic stroke,3 peripheral arterial disease,2 deep vein thrombosis,6 and pulmonary embolism6 (Supplementary material online, Figure S1). However, this and a previous MR study based on another study sample (n = 1545 cases of intracerebral haemorrhage and 1481 controls)25 found limited evidence of a causal association between smoking and intracerebral haemorrhage and subarachnoid haemorrhage. These findings are in contrast to those of a large prospective study and meta-analysis which showed that smoking was associated with increased risk of both intracerebral haemorrhage and subarachnoid haemorrhage3 (Supplementary material online, Figure S1). The present and the prior MR study25 may have been unable to observe a statistically significant association due to small number of cases of haemorrhagic stroke. With regard to ischaemic stroke subtypes, smoking has been shown to increase the risk of large artery and small vessel stroke but not cardioembolic stroke.25 Data on ischaemic stroke subtypes were not available in UK Biobank.

In contrast to abdominal aortic aneurysm, we found no evidence of a causal association between smoking and thoracic aortic aneurysm but owing to the limited number of cases, we cannot rule out that we may have overlooked a weak association. Observational studies of the association between smoking and thoracic aortic aneurysm are scarce. No association between smoking and risk of rupture or dissection of thoracic aortic aneurysm was observed in a cohort of thoracic aortic aneurysm patients without history of aortic dissection.11 However, smoking was associated with a 2.2-fold higher risk of thoracic aortic aneurysm in a cohort of 30 447 adults, including 45 thoracic aortic aneurysm cases.12 Thus, the role of smoking for thoracic aortic aneurysm requires further study.

Observational studies of smoking in relation to risk of aortic valve stenosis7–9 and atrial fibrillation10 are limited or inconsistent. One large9 and two small prospective cohort studies7  ,  8 found a 30%9 to almost two-fold7 increased risk of aortic valve stenosis associated with smoking, and a meta-analysis found an overall 32% increased risk of atrial fibrillation for current vs. never smokers.10 This MR study found no association between smoking initiation and aortic valve stenosis after adjustment for body mass index, which is a strong risk factor for aortic valve stenosis.32 Likewise, no consistent association was observed between smoking initiation and atrial fibrillation.

Potential mechanisms

Cigarette smoke can damage the cardiovascular system through its content of oxidant gases (e.g. oxides of nitrogen and free radicals) and other toxic substances, which may increase the risk of CVD through lipid oxidation, endothelial dysfunction, inflammation platelet activation, thrombogenesis, and augmented coagulability.33 Furthermore, nicotine may provoke acute cardiovascular events by increasing myocardial contractility and vasoconstriction, leading to increased myocardial work and oxygen demand as well as reduced coronary and cerebral blood flow.33 Nicotine exposure may further promote aneurysmal rupture through actions on vascular smooth muscle cell nicotinic acetylcholine receptors containing α7 subunits.34 Smoking might also affect CVD risk through antioestrogenic effects, observed in both women and men.35–37

Strength and limitations

This study has several strengths. First, the MR design reduces the possibility that the observed associations were biased by reverse causality and confounding. Second, the genetic instrument for smoking initiation comprised multiple SNPs robustly associated with smoking, thereby providing a strong genetic instrument. Third, the analyses included a large sample size and generally large number of CVD events, and therefore, the statistical power was high in most analyses. Fourth, by restricting the study sample to individuals of European-descent in UK Biobank, population stratification bias was minimized.

This study also has several limitations. First, the possibility that the smoking-related SNPs affect CVD outcomes through other causal pathways than through smoking exposure cannot be entirely ruled out. Nonetheless, in never smokers the associations of genetic predisposition to smoking initiation with the CVD outcomes were generally null (except for arterial hypertension) in the crude or multivariable MR analysis. The weak associations for some CVD outcomes may be due to exposure to cigarette smoke amongst never smokers or to misclassification of smoking status (ever smokers incorrectly reported never smoked regularly) as self-reported smoking prevalence is known to be lower than directly measured values of cotinine levels.38

Second, the causal association between smoking cessation and CVD could not be assessed because identified SNPs for smoking cessation are few and explains a small proportion of the phenotypic variance (0.1%).20 To address those limitations, we conducted a complementary analysis of lifetime smoking index, which takes into account smoking duration, heaviness, and cessation,22 and observed similar results to those based on the smoking initiation phenotype.

A third limitation is that participants of the UK Biobank were included in both the exposure and outcome datasets (about 30% participant overlap) in the primary analyses. This overlap may have introduced some bias in the causal estimates in the direction of the observational association between smoking and CVD risk.39 However, the genetic variants are reasonably strongly associated with the exposure (F statistic 78.5), meaning that bias due to sample overlap is reasonably small. The associations of genetic predisposition to smoking initiation with coronary artery disease, ischaemic stroke, atrial fibrillation, and venous thromboembolism were stronger in UK Biobank than in non-overlapping sets of individuals included in the consortia. This might be related to more consistent case definitions in UK Biobank, population mixture in the consortia (as the genetic variants were chosen based on their associations with smoking status in Europeans, we may expect associations to be attenuated in a mixed ethnicity population), or differences in smoking status and prevalence of other risk factors in different populations. For example, the MVP cohort consists of over 90% men and the prevalence of ever smoking (about 75%) and CVD risk factors, such as Type 2 diabetes, hyperlipidaemia, and obesity, is much higher in the MVP cohort than in the UK Biobank.19 As environmental influences on smoking are stronger in this cohort, genetic influences on smoking are likely to be weaker.

Finally, the power was insufficient for analyses of the association of genetic predisposition to smoking with CVD in individuals of non-European ancestry. Genome-wide associations studies assessing the genetics of smoking in non-Europeans, as well as large MR studies of smoking and CVD risk in other populations, are warranted.

Conclusions

This study supports a causal association between smoking and a broad range of CVDs, in particular, coronary artery disease, heart failure, abdominal aortic aneurysm, ischaemic stroke, transient ischaemic attack, peripheral arterial disease, and arterial hypertension. These findings add to the level of evidence that smoking is a causal risk factor for CVD.40

Supplementary Material

ehaa193_Supplementary_Data
Click here for additional data file. (395.2KB, zip)

Acknowledgements

This research was conducted using the UK Biobank study under Application Number 29202.

Funding

This work was supported by funding from the Swedish Research Council (Vetenskapsrådet; 2019-00977 to S.C.L.), the Swedish Heart-Lung Foundation (Hjärt-Lungfonden; 20190247 to S.C.L.), the Swedish Research Council for Health, Working Life and Welfare (Forte; 2018-00123 to S.C.L.), and the United States Department of Veterans Affairs Office of Research and Development, Million Veteran Program Grant # MVP003 (I01-BX003362). S.M.D. was supported by the VA award IK2-CX001780. S.B. was supported by a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (204623/Z/16/Z). This publication does not represent the views of the Department of Veterans Affairs or the United States Government.

Conflict of interest: Dr Damrauer reported receiving grants from the U.S. Department of Veterans Affairs and Renalytix AI plc outside the submitted work. No other disclosures were reported.

Contributor Information

Susanna C Larsson, Department of Surgical Sciences, Uppsala University, Uppsala 75185, Sweden; Unit of Cardiovascular and Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm 17177, Sweden.

Amy M Mason, Department of Public Health and Primary Care, University of Cambridge, Cambridge CB1 8RN, UK.

Magnus Bäck, Department of Medicine, Center for Molecular Medicine, Karolinska Institutet, Stockholm 17177, Sweden; Division of Valvular and Coronary Disease, Heart and Vascular Theme, Karolinska University Hospital, Stockholm 14186, Sweden.

Derek Klarin, Boston VA Healthcare System, Boston, MA 02132-4927, USA; Center for Genomic Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, USA; Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Division of Vascular Surgery and Endovascular Therapy, University of Florida College of Medicine, Gainesville, FL 32610, USA.

Scott M Damrauer, Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA 19104, USA; Department of Surgery, Perlman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA.

Karl Michaëlsson, Department of Surgical Sciences, Uppsala University, Uppsala 75185, Sweden.

Stephen Burgess, Department of Public Health and Primary Care, University of Cambridge, Cambridge CB1 8RN, UK; MRC Biostatistics Unit, University of Cambridge, Cambridge CB20SR, UK.

References

  • 1. Mons  U, Muezzinler  A, Gellert  C, Schottker  B, Abnet  CC, Bobak  M, de Groot  L, Freedman  ND, Jansen  E, Kee  F, Kromhout  D, Kuulasmaa  K, Laatikainen  T, O'Doherty  MG, Bueno-de-Mesquita  B, Orfanos  P, Peters  A, van der Schouw  YT, Wilsgaard  T, Wolk  A, Trichopoulou  A, Boffetta  P, Brenner  H; CHANCES Consortium. Impact of smoking and smoking cessation on cardiovascular events and mortality among older adults: meta-analysis of individual participant data from prospective cohort studies of the CHANCES consortium. BMJ  2015;350:h1551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Lu  L, Mackay  DF, Pell  JP.  Meta-analysis of the association between cigarette smoking and peripheral arterial disease. Heart  2014;100:414–423. [DOI] [PubMed] [Google Scholar]
  • 3. Price  AJ, Wright  FL, Green  J, Balkwill  A, Kan  SW, Yang  TO, Floud  S, Kroll  ME, Simpson  R, Sudlow  CLM, Beral  V, Reeves  GK.  Differences in risk factors for 3 types of stroke: UK prospective study and meta-analyses. Neurology  2018;90:e298–e306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Aune  D, Schlesinger  S, Norat  T, Riboli  E.  Tobacco smoking and the risk of heart failure: a systematic review and meta-analysis of prospective studies. Eur J Prev Cardiol  2019;26:279–288. [DOI] [PubMed] [Google Scholar]
  • 5. Aune  D, Schlesinger  S, Norat  T, Riboli  E.  Tobacco smoking and the risk of abdominal aortic aneurysm: a systematic review and meta-analysis of prospective studies. Sci Rep  2018;8:14786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Gregson  J, Kaptoge  S, Bolton  T, Pennells  L, Willeit  P, Burgess  S, Bell  S, Sweeting  M, Rimm  EB, Kabrhel  C, Zoller  B, Assmann  G, Gudnason  V, Folsom  AR, Arndt  V, Fletcher  A, Norman  PE, Nordestgaard  BG, Kitamura  A, Mahmoodi  BK, Whincup  PH, Knuiman  M, Salomaa  V, Meisinger  C, Koenig  W, Kavousi  M, Volzke  H, Cooper  JA, Ninomiya  T, Casiglia  E, Rodriguez  B, Ben-Shlomo  Y, Despres  JP, Simons  L, Barrett-Connor  E, Bjorkelund  C, Notdurfter  M, Kromhout  D, Price  J, Sutherland  SE, Sundstrom  J, Kauhanen  J, Gallacher  J, Beulens  JWJ, Dankner  R, Cooper  C, Giampaoli  S, Deen  JF, Gomez de la Camara  A, Kuller  LH, Rosengren  A, Svensson  PJ, Nagel  D, Crespo  CJ, Brenner  H, Albertorio-Diaz  JR, Atkins  R, Brunner  EJ, Shipley  M, Njolstad  I, Lawlor  DA, van der Schouw  YT, Selmer  RM, Trevisan  M, Verschuren  WMM, Greenland  P, Wassertheil-Smoller  S, Lowe  GDO, Wood  AM, Butterworth  AS, Thompson  SG, Danesh  J, Di Angelantonio  E, Meade  T, Emerging Risk Factors Collaboration. Cardiovascular risk factors associated with venous thromboembolism. JAMA Cardiol  2019;4:163–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Martinsson  A, Ostling  G, Persson  M, Sundquist  K, Andersson  C, Melander  O, Engstrom  G, Hedblad  B, Smith  JG.  Carotid plaque, intima-media thickness, and incident aortic stenosis: a prospective cohort study. Arterioscler Thromb Vasc Biol  2014;34:2343–2348. [DOI] [PubMed] [Google Scholar]
  • 8. Eveborn  GW, Schirmer  H, Lunde  P, Heggelund  G, Hansen  JB, Rasmussen  K.  Assessment of risk factors for developing incident aortic stenosis: the Tromso Study. Eur J Epidemiol  2014;29:567–575. [DOI] [PubMed] [Google Scholar]
  • 9. Larsson  SC, Wolk  A, Back  M.  Alcohol consumption, cigarette smoking and incidence of aortic valve stenosis. J Intern Med  2017;282:332–339. [DOI] [PubMed] [Google Scholar]
  • 10. Aune  D, Schlesinger  S, Norat  T, Riboli  E.  Tobacco smoking and the risk of atrial fibrillation: a systematic review and meta-analysis of prospective studies. Eur J Prev Cardiol  2018;25:1437–1451. [DOI] [PubMed] [Google Scholar]
  • 11. Kim  JB, Kim  K, Lindsay  ME, MacGillivray  T, Isselbacher  EM, Cambria  RP, Sundt  TM  3rd. Risk of rupture or dissection in descending thoracic aortic aneurysm. Circulation  2015;132:1620–1629. [DOI] [PubMed] [Google Scholar]
  • 12. Landenhed  M, Engstrom  G, Gottsater  A, Caulfield  MP, Hedblad  B, Newton-Cheh  C, Melander  O, Smith  JG.  Risk profiles for aortic dissection and ruptured or surgically treated aneurysms: a prospective cohort study. J Am Heart Assoc  2015;4:e001513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Rose  G, Colwell  L.  Randomised controlled trial of anti-smoking advice: final (20 year) results. J Epidemiol Community Health  1992;46:75–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Multiple risk factor intervention trial. Risk factor changes and mortality results. Multiple Risk Factor Intervention Trial Research Group. JAMA  1982;248:1465–1477. [PubMed] [Google Scholar]
  • 15. Davey Smith  G, Ebrahim  S.  ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease?  Int J Epidemiol  2003;32:1–22. [DOI] [PubMed] [Google Scholar]
  • 16. Burgess  S, Thompson  SG.  Mendelian Randomization: Methods for Using Genetic Variants in Causal Estimation. London, UK: Chapman and Hall/CRC Press; 2015. [Google Scholar]
  • 17. Nikpay  M, Goel  A, Won  HH, Hall  LM, Willenborg  C, Kanoni  S, Saleheen  D, Kyriakou  T, Nelson  CP, Hopewell  JC, Webb  TR, Zeng  L, Dehghan  A, Alver  M, Armasu  SM, Auro  K, Bjonnes  A, Chasman  DI, Chen  S, Ford  I, Franceschini  N, Gieger  C, Grace  C, Gustafsson  S, Huang  J, Hwang  SJ, Kim  YK, Kleber  ME, Lau  KW, Lu  X, Lu  Y, Lyytikainen  LP, Mihailov  E, Morrison  AC, Pervjakova  N, Qu  L, Rose  LM, Salfati  E, Saxena  R, Scholz  M, Smith  AV, Tikkanen  E, Uitterlinden  A, Yang  X, Zhang  W, Zhao  W, de Andrade  M, de Vries  PS, van Zuydam  NR, Anand  SS, Bertram  L, Beutner  F, Dedoussis  G, Frossard  P, Gauguier  D, Goodall  AH, Gottesman  O, Haber  M, Han  BG, Huang  J, Jalilzadeh  S, Kessler  T, Konig  IR, Lannfelt  L, Lieb  W, Lind  L, Lindgren  CM, Lokki  ML, Magnusson  PK, Mallick  NH, Mehra  N, Meitinger  T, Memon  FU, Morris  AP, Nieminen  MS, Pedersen  NL, Peters  A, Rallidis  LS, Rasheed  A, Samuel  M, Shah  SH, Sinisalo  J, Stirrups  KE, Trompet  S, Wang  L, Zaman  KS, Ardissino  D, Boerwinkle  E, Borecki  IB, Bottinger  EP, Buring  JE, Chambers  JC, Collins  R, Cupples  LA, Danesh  J, Demuth  I, Elosua  R, Epstein  SE, Esko  T, Feitosa  MF, Franco  OH, Franzosi  MG, Granger  CB, Gu  D, Gudnason  V, Hall  AS, Hamsten  A, Harris  TB, Hazen  SL, Hengstenberg  C, Hofman  A, Ingelsson  E, Iribarren  C, Jukema  JW, Karhunen  PJ, Kim  BJ, Kooner  JS, Kullo  IJ, Lehtimaki  T, Loos  RJ, Melander  O, Metspalu  A, Marz  W, Palmer  CN, Perola  M, Quertermous  T, Rader  DJ, Ridker  PM, Ripatti  S, Roberts  R, Salomaa  V, Sanghera  DK, Schwartz  SM, Seedorf  U, Stewart  AF, Stott  DJ, Thiery  J, Zalloua  PA, O'Donnell  CJ, Reilly  MP, Assimes  TL, Thompson  JR, Erdmann  J, Clarke  R, Watkins  H, Kathiresan  S, McPherson  R, Deloukas  P, Schunkert  H, Samani  NJ, Farrall  M.  A comprehensive 1,000 Genomes-based genome-wide association meta-analysis of coronary artery disease. Nat Genet  2015;47:1121–1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Christophersen  IE, Rienstra  M, Roselli  C, Yin  X, Geelhoed  B, Barnard  J, Lin  H, Arking  DE, Smith  AV, Albert  CM, Chaffin  M, Tucker  NR, Li  M, Klarin  D, Bihlmeyer  NA, Low  S-K, Weeke  PE, Müller-Nurasyid  M, Smith  JG, Brody  JA, Niemeijer  MN, Dörr  M, Trompet  S, Huffman  J, Gustafsson  S, Schurmann  C, Kleber  ME, Lyytikäinen  L-P, Seppälä  I, Malik  R, Horimoto  ARVR, Perez  M, Sinisalo  J, Aeschbacher  S, Thériault  S, Yao  J, Radmanesh  F, Weiss  S, Teumer  A, Choi  SH, Weng  L-C, Clauss  S, Deo  R, Rader  DJ, Shah  SH, Sun  A, Hopewell  JC, Debette  S, Chauhan  G, Yang  Q, Worrall  BB, Paré  G, Kamatani  Y, Hagemeijer  YP, Verweij  N, Siland  JE, Kubo  M, Smith  JD, Van Wagoner  DR, Bis  JC, Perz  S, Psaty  BM, Ridker  PM, Magnani  JW, Harris  TB, Launer  LJ, Shoemaker  MB, Padmanabhan  S, Haessler  J, Bartz  TM, Waldenberger  M, Lichtner  P, Arendt  M, Krieger  JE, Kähönen  M, Risch  L, Mansur  AJ, Peters  A, Smith  BH, Lind  L, Scott  SA, Lu  Y, Bottinger  EB, Hernesniemi  J, Lindgren  CM, Wong  JA, Huang  J, Eskola  M, Morris  AP, Ford  I, Reiner  AP, Delgado  G, Chen  LY, Chen  Y-DI, Sandhu  RK, Li  M, Boerwinkle  E, Eisele  L, Lannfelt  L, Rost  N, Anderson  CD, Taylor  KD, Campbell  A, Magnusson  PK, Porteous  D, Hocking  LJ, Vlachopoulou  E, Pedersen  NL, Nikus  K, Orho-Melander  M, Hamsten  A, Heeringa  J, Denny  JC, Kriebel  J, Darbar  D, Newton-Cheh  C, Shaffer  C, Macfarlane  PW, Heilmann-Heimbach  S, Almgren  P, Huang  PL, Sotoodehnia  N, Soliman  EZ, Uitterlinden  AG, Hofman  A, Franco  OH, Völker  U, Jöckel  K-H, Sinner  MF, Lin  HJ, Guo  X, Dichgans  M, Ingelsson  E, Kooperberg  C, Melander  O, Loos  RJF, Laurikka  J, Conen  D, Rosand  J, van der Harst  P, Lokki  M-L, Kathiresan  S, Pereira  A, Jukema  JW, Hayward  C, Rotter  JI, März  W, Lehtimäki  T, Stricker  BH, Chung  MK, Felix  SB, Gudnason  V, Alonso  A, Roden  DM, Kääb  S, Chasman  DI, Heckbert  SR, Benjamin  EJ, Tanaka  T, Lunetta  KL, Lubitz  SA, Ellinor  PT; AFGen Consortium. Large-scale analyses of common and rare variants identify 12 new loci associated with atrial fibrillation. Nat Genet  2017;49:946–952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Klarin  D, Busenkell  E, Judy  R, Lynch  J, Levin  M, Haessler  J, Aragam  K, Chaffin  M, Haas  M, Lindstrom  S, Assimes  TL, Huang  J, Min Lee  K, Shao  Q, Huffman  JE, Kabrhel  C, Huang  Y, Sun  YV, Vujkovic  M, Saleheen  D, Miller  DR, Reaven  P, DuVall  S, Boden  WE, Pyarajan  S, Reiner  AP, Tregouet  DA, Henke  P, Kooperberg  C, Gaziano  JM, Concato  J, Rader  DJ, Cho  K, Chang  KM, Wilson  PWF, Smith  NL, O'Donnell  CJ, Tsao  PS, Kathiresan  S, Obi  A, Damrauer  SM, Natarajan  P; INVENT Consortium; Veterans Affairs’ Million Veteran Program. Genome-wide association analysis of venous thromboembolism identifies new risk loci and genetic overlap with arterial vascular disease. Nat Genet  2019;51:1574–1579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Liu  M, Jiang  Y, Wedow  R, Li  Y, Brazel  DM, Chen  F, Datta  G, Davila-Velderrain  J, McGuire  D, Tian  C, Zhan  X23andMe Research Team; HUNT All-In PsychiatryChoquet  H, Docherty  AR, Faul  JD, Foerster  JR, Fritsche  LG, Gabrielsen  ME, Gordon  SD, Haessler  J, Hottenga  JJ, Huang  H, Jang  SK, Jansen  PR, Ling  Y, Magi  R, Matoba  N, McMahon  G, Mulas  A, Orru  V, Palviainen  T, Pandit  A, Reginsson  GW, Skogholt  AH, Smith  JA, Taylor  AE, Turman  C, Willemsen  G, Young  H, Young  KA, Zajac  GJM, Zhao  W, Zhou  W, Bjornsdottir  G, Boardman  JD, Boehnke  M, Boomsma  DI, Chen  C, Cucca  F, Davies  GE, Eaton  CB, Ehringer  MA, Esko  T, Fiorillo  E, Gillespie  NA, Gudbjartsson  DF, Haller  T, Harris  KM, Heath  AC, Hewitt  JK, Hickie  IB, Hokanson  JE, Hopfer  CJ, Hunter  DJ, Iacono  WG, Johnson  EO, Kamatani  Y, Kardia  SLR, Keller  MC, Kellis  M, Kooperberg  C, Kraft  P, Krauter  KS, Laakso  M, Lind  PA, Loukola  A, Lutz  SM, Madden  PAF, Martin  NG, McGue  M, McQueen  MB, Medland  SE, Metspalu  A, Mohlke  KL, Nielsen  JB, Okada  Y, Peters  U, Polderman  TJC, Posthuma  D, Reiner  AP, Rice  JP, Rimm  E, Rose  RJ, Runarsdottir  V, Stallings  MC, Stancakova  A, Stefansson  H, Thai  KK, Tindle  HA, Tyrfingsson  T, Wall  TL, Weir  DR, Weisner  C, Whitfield  JB, Winsvold  BS, Yin  J, Zuccolo  L, Bierut  LJ, Hveem  K, Lee  JJ, Munafo  MR, Saccone  NL, Willer  CJ, Cornelis  MC, David  SP, Hinds  DA, Jorgenson  E, Kaprio  J, Stitzel  JA, Stefansson  K, Thorgeirsson  TE, Abecasis  G, Liu  DJ, Vrieze  S.  Association studies of up to 1.2 million individuals yield new insights into the genetic etiology of tobacco and alcohol use. Nat Genet  2019;51:237–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Machiela  MJ, Chanock  SJ.  LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics  2015;31:3555–3557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Wootton  RE, Richmond  RC, Stuijfzand  BG, Lawn  RB, Sallis  HM, Taylor  GMJ, Jones  HJ, Zammit  S, Smith Munafo  G, Munafò  MR.  Evidence for causal effects of lifetime smoking on risk for depression and schizophrenia: a Mendelian randomisation study. Psychol Med  2019;6:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Sudlow  C, Gallacher  J, Allen  N, Beral  V, Burton  P, Danesh  J, Downey  P, Elliott  P, Green  J, Landray  M, Liu  B, Matthews  P, Ong  G, Pell  J, Silman  A, Young  A, Sprosen  T, Peakman  T, Collins  R.  UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med  2015;12:e1001779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Malik  R, Chauhan  G, Traylor  M, Sargurupremraj  M, Okada  Y, Mishra  A, Rutten-Jacobs  L, Giese  AK, van der Laan  SW, Gretarsdottir  S, Anderson  CD, Chong  M, Adams  HHH, Ago  T, Almgren  P, Amouyel  P, Ay  H, Bartz  TM, Benavente  OR, Bevan  S, Boncoraglio  GB, Brown  RD  Jr, Butterworth  AS, Carrera  C, Carty  CL, Chasman  DI, Chen  WM, Cole  JW, Correa  A, Cotlarciuc  I, Cruchaga  C, Danesh  J, de Bakker  PIW, DeStefano  AL, den Hoed  M, Duan  Q, Engelter  ST, Falcone  GJ, Gottesman  RF, Grewal  RP, Gudnason  V, Gustafsson  S, Haessler  J, Harris  TB, Hassan  A, Havulinna  AS, Heckbert  SR, Holliday  EG, Howard  G, Hsu  FC, Hyacinth  HI, Ikram  MA, Ingelsson  E, Irvin  MR, Jian  X, Jiménez-Conde  J, Johnson  JA, Jukema  JW, Kanai  M, Keene  KL, Kissela  BM, Kleindorfer  DO, Kooperberg  C, Kubo  M, Lange  LA, Langefeld  CD, Langenberg  C, Launer  LJ, Lee  JM, Lemmens  R, Leys  D, Lewis  CM, Lin  WY, Lindgren  AG, Lorentzen  E, Magnusson  PK, Maguire  J, Manichaikul  A, McArdle  PF, Meschia  JF, Mitchell  BD, Mosley  TH, Nalls  MA, Ninomiya  T, O'Donnell  MJ, Psaty  BM, Pulit  SL, Rannikmäe  K, Reiner  AP, Rexrode  KM, Rice  K, Rich  SS, Ridker  PM, Rost  NS, Rothwell  PM, Rotter  JI, Rundek  T, Sacco  RL, Sakaue  S, Sale  MM, Salomaa  V, Sapkota  BR, Schmidt  R, Schmidt  CO, Schminke  U, Sharma  P, Slowik  A, Sudlow  CLM, Tanislav  C, Tatlisumak  T, Taylor  KD, Thijs  VNS, Thorleifsson  G, Thorsteinsdottir  U, Tiedt  S, Trompet  S, Tzourio  C, van Duijn  CM, Walters  M, Wareham  NJ, Wassertheil-Smoller  S, Wilson  JG, Wiggins  KL, Yang  Q, Yusuf  SAFGen Consortium; Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium; International Genomics of Blood Pressure (iGEN-BP) Consortium; INVENT Consortium; STARNETBis  JC, Pastinen  T, Ruusalepp  A, Schadt  EE, Koplev  S, Björkegren  JLM, Codoni  V, Civelek  M, Smith  NL, Trégouët  DA, Christophersen  IE, Roselli  C, Lubitz  SA, Ellinor  PT, Tai  ES, Kooner  JS, Kato  N, He  J, van der Harst  P, Elliott  P, Chambers  JC, Takeuchi  F, Johnson  ADBioBank Japan Cooperative Hospital Group; COMPASS Consortium; EPIC-CVD Consortium; EPIC-InterAct Consortium; International Stroke Genetics Consortium (ISGC); METASTROKE Consortium; Neurology Working Group of the CHARGE Consortium; NINDS Stroke Genetics Network (SiGN); UK Young Lacunar DNA Study; MEGASTROKE ConsortiumSanghera  DK, Melander  O, Jern  C, Strbian  D, Fernandez-Cadenas  I, Longstreth  WT  Jr, Rolfs  A, Hata  J, Woo  D, Rosand  J, Pare  G, Hopewell  JC, Saleheen  D, Stefansson  K, Worrall  BB, Kittner  SJ, Seshadri  S, Fornage  M, Markus  HS, Howson  JMM, Kamatani  Y, Debette  S, Dichgans  M.  Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet  2018;50:524–537.29531354 [Google Scholar]
  • 25. Larsson  SC, Burgess  S, Michaëlsson  K.  Smoking and stroke: a Mendelian randomization study. Ann Neurol  2019;86:468–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Higgins  JP, Thompson  SG.  Quantifying heterogeneity in a meta-analysis. Stat Med  2002;21:1539–1558. [DOI] [PubMed] [Google Scholar]
  • 27. Burgess  S, Thompson  SG.  Multivariable Mendelian randomization: the use of pleiotropic genetic variants to estimate causal effects. Am J Epidemiol  2015;181:251–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Ntalla  I, Kanoni  S, Zeng  L, Giannakopoulou  O, Danesh  J, Watkins  H, Samani  NJ, Deloukas  P, Schunkert  H, Group  U.  Genetic risk score for coronary disease identifies predispositions to cardiovascular and noncardiovascular diseases. J Am Coll Cardiol  2019;73:2932–2942. [DOI] [PubMed] [Google Scholar]
  • 29. Burgess  S, Bowden  J, Fall  T, Ingelsson  E, Thompson  SG.  Sensitivity analyses for robust causal inference from Mendelian randomization analyses with multiple genetic variants. Epidemiology  2017;28:30–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Spiller  W, Davies  NM, Palmer  TM.  Software application profile: mrrobust—a tool for performing two-sample summary Mendelian randomization analyses. Int J Epidemiol  2019;48:684–690. [Google Scholar]
  • 31. Yavorska  OO, Burgess  S.  MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. Int J Epidemiol  2017;46:1734–1739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Larsson  SC, Bäck  M, Rees  JMB, Mason  AM, Burgess  S.  Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study. Eur Heart J  2020;41:221–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Rigotti  NA, Clair  C.  Managing tobacco use: the neglected cardiovascular disease risk factor. Eur Heart J  2013;34:3259–3267. [DOI] [PubMed] [Google Scholar]
  • 34. Kamio  Y, Miyamoto  T, Kimura  T, Mitsui  K, Furukawa  H, Zhang  D, Yokosuka  K, Korai  M, Kudo  D, Lukas  RJ, Lawton  MT, Hashimoto  T.  Roles of nicotine in the development of intracranial aneurysm rupture. Stroke  2018;49:2445–2452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Baron  JA, La Vecchia  C, Levi  F.  The antiestrogenic effect of cigarette smoking in women. Am J Obstet Gynecol  1990;162:502–514. [DOI] [PubMed] [Google Scholar]
  • 36. Michnovicz  JJ, Hershcopf  RJ, Haley  NJ, Bradlow  HL, Fishman  J.  Cigarette smoking alters hepatic estrogen metabolism in men: implications for atherosclerosis. Metabolism  1989;38:537–541. [DOI] [PubMed] [Google Scholar]
  • 37. Michnovicz  JJ, Hershcopf  RJ, Naganuma  H, Bradlow  HL, Fishman  J.  Increased 2-hydroxylation of estradiol as a possible mechanism for the anti-estrogenic effect of cigarette smoking. N Engl J Med  1986;315:1305–1309. [DOI] [PubMed] [Google Scholar]
  • 38. Connor Gorber  S, Schofield-Hurwitz  S, Hardt  J, Levasseur  G, Tremblay  M.  The accuracy of self-reported smoking: a systematic review of the relationship between self-reported and cotinine-assessed smoking status. Nicotine Tob Res  2009;11:12–24. [DOI] [PubMed] [Google Scholar]
  • 39. Burgess  S, Davies  NM, Thompson  SG.  Bias due to participant overlap in two-sample Mendelian randomization. Genet Epidemiol  2016;40:597–608. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ehaa193_Supplementary_Data
Click here for additional data file. (395.2KB, zip)

Articles from European Heart Journal are provided here courtesy of Oxford University Press

RESOURCES