gms | German Medical Science

50. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds)
12. Jahrestagung der Deutschen Arbeitsgemeinschaft für Epidemiologie (dae)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie
Deutsche Arbeitsgemeinschaft für Epidemiologie

12. bis 15.09.2005, Freiburg im Breisgau

Smoking Behaviour and Subclinical Atherosclerosis – Results from the Heinz Nixdorf Recall Study

Meeting Abstract

  • K.-H. Jöckel - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Universitätsklinikum Essen, Essen
  • N. Lehmann - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Universitätsklinikum Essen, Essen
  • B. Jaeger - Abt. für Kardiologie, Universitätsklinikum Essen, Essen
  • S. Moebus - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Universitätsklinikum Essen, Essen
  • A. Schmermund - Abt. für Kardiologie, Universitätsklinikum Essen, Essen
  • S. Möhlenkamp - Abt. für Kardiologie, Universitätsklinikum Essen, Essen
  • N. Dragano - Institut für Medizinische Soziologie, Universität Düsseldorf, Düsseldorf
  • A. Stang - Institut für Medizinische Epidemiologie, Biometrie und Informatik, Universität Halle, Halle
  • R. Erbel - Abt. für Kardiologie, Universitätsklinikum Essen, Essen

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. Deutsche Arbeitsgemeinschaft für Epidemiologie. 50. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 12. Jahrestagung der Deutschen Arbeitsgemeinschaft für Epidemiologie. Freiburg im Breisgau, 12.-15.09.2005. Düsseldorf, Köln: German Medical Science; 2005. Doc05gmds374

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter:

Veröffentlicht: 8. September 2005

© 2005 Jöckel et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Introduction and study aims

Smoking still is a common habit worldwide with major negative impact on public health, being causatively linked to the development of various diseases like myocardial infarction, stroke, coronary artery disease (CAD), peripheral artery disease and lung cancer. It largely contributes to cardiovascular disease, the latter being the leading cause of death at least in the industrialized nations. One key issue of preventive medicine is to identify asymptomatic individuals at high risk already in the stage of subclinical atherosclerosis, because 50% of coronary heart disease deaths occur in that stage as sudden cardiac deaths. Therefore we investigated the relationship between smoking habits and subclinical atherosclerosis in a large cross-sectional study.

Subclinical atherosclerosis can be visualized and quantified noninvasively (as coronary artery calcification or CAC score) by electron-beam computed tomography (EBCT). CAC burden correlates with atherosclerotic plaque area, and an increased quantity of CAC closely correlates with clinical manifestation of CAD.

Assessment of CAC in the present population-based study allows us to analyze quantitatively its relation with smoking behaviour, accounting for the classical CAD risk factors as included in the Framingham and PROCAM algorithms. Further risk factors analyzed comprise body mass index, fibrinogen, C-reactive protein and lipoprotein(a).

Material and methods

The Heinz Nixdorf Recall Study is an ongoing study to clarify the predictive value of CAC for future cardiac events [1]. The study population is a random sample derived from mandatory citizen registries in three German cities (Bochum, Essen, Mülheim/Ruhr) in the Ruhr area. Participants were neither physician- nor self-referred.

The Heinz Nixdorf Recall Study is funded by the Heinz Nixdorf Foundation. The study was reviewed and approved by the institutional review boards. It was planned and conducted according to Good Epidemiological Practice (GEP) as put forward by the German Epidemiological Association (DAE). This included a strict internal and external quality control. Study procedures were quality certificated according to DIN EN ISO 9001:2000.

A total of 4814 participants aged 45-75 years were recruited in the period December 2000 to August 2003. The baseline response proportion of the cohort study expressed as recruitment efficacy proportion was 55.8% based on subjects without a manifest coronary artery disease. We excluded subjects with a history of coronary

artery disease (n=327) or stroke (n=135). At the time of the present analysis, there were 205 test persons with missing CAC and 124 with missing or inconsistent data on smoking. This results in an effective sample size of 4078 subjects.

All study participants underwent a standardized computer assisted interview covering medical history, sporting activities, social background, working environment and nutrition (about 300 parameters in total). Smoking behaviour was assessed by questions on smoking status with present smoking defined as a history of regular smoking in the past year, age at start and (if former smoker) year of end of smoking, and the number of cigarettes or pipes/cigars smoked daily. The intake of blood pressure lowering or lipid lowering medication was documented as well as a physician’s diagnosis of diabetes. Blood pressure and laboratory parameters were assessed according to predefined standards, the amount of CAC was measured following the Agatston method.

For the statistical analysis we used log(CAC score +1) as the primary study variable to account for the skewness of the distribution. Multiple regression analyses were performed using SAS version 8.2 procedures GLM and REG for the full study population and in the subgroups of never, former and present smokers.


In the given age range the level of CAC is much higher in men than in women, roughly by a factor of five. With increasing age CAC increases exponentially, never smokers showing lower scores than ex-smokers, ex-smokers lower than present smokers. A formal regression analysis within smoking groups confirms this first impression by showing a reasonable fit with R2 values of 0.21, 0.30, 0.26 for never, former, present smokers, respectively. If we include time since quitting and dose into the model for ex-smokers and dose for present smokers the improvement in fit is statistically significant for these models.

Fig. 1 [Fig. 1] displays the predicted calcium score depending on age for a male non-smoker, an ex-smoker having quit smoking 30 cigarettes per day 20 years ago and a present smoker smoking 30 cigarettes per day. This figure gives rise to the rule of thumb that with respect to coronary calcium a male ex-smoker is 5 years “older” than a non-smoker and that a current heavy smoker even exceeds his non-smoking counterpart by 10 years of age. For females, current heavy smoking is associated with an increase in CAC equivalent to a 10 year aging effect as well, but in a female former smoking relates to a lower increase in CAC comparable to 2.5 additional years of age.

Including cardiac risk factors in the regression analysis leads in the full study population to a fit with R2=0.32, where age, gender, former vs. never smoking, present vs. never smoking, antihypertensive medication, lipid lowering medication, diabetes, body mass index, blood pressure amplitude, low density lipoprotein, fibrinogen and lipoprotein(a) are independently predictive. In the smoking behaviour subgroups slightly varying dependency patterns emerge.


First results of our study show a strong positive association between smoking and coronary calcium score as measured by EBCT. For a heavy smoker, to smoke or never having smoked makes an approximate 10 year difference in “coronary artery age” for both men and women with the difference being consistent over time in all age groups. At first glance surprisingly, a history of heavy former smoking – even when this was twenty years ago – already amounts to a 5 year difference in terms of “artery age” in men and 2.5 years in women. This finding is of special interest concerning public health effects of the policy of tobacco ban. Gender differences in CAC should be seen in conjunction with the fact that the average Framingham risk in this population is 2 times higher in males than in females. The quantitative role of risk factors – classical and novel – will be discussed.


Schmermund A, Möhlenkamp S, Stang A. Assessment of clinically silent atherosclerotic disease and established and novel risk factors for predicting myocardial infarction and cardiac death in healthy middle-aged subjects: Rationale and design of the Heinz Nixdorf Recall Study. Am Heart J 2002; 144: 212-18.