Article Text

Download PDFPDF
Association between anthropometric obesity measures and coronary artery disease: a cross-sectional survey of 16 657 subjects from 444 Polish cities
  1. B M Kaess1,2,
  2. J Jozwiak3,4,
  3. M Mastej4,5,
  4. W Lukas3,
  5. W Grzeszczak6,
  6. A Windak7,
  7. W Piwowarska8,
  8. A Tykarski9,
  9. E Konduracka8,
  10. K Rygiel3,
  11. A Manasar4,
  12. N J Samani1,
  13. M Tomaszewski1
  1. 1
    Department of Cardiovascular Sciences, University of Leicester, Leicester, UK
  2. 2
    Klinik und Poliklinik für Innere Medizin II, University of Regensburg, Regensburg, Germany
  3. 3
    Department of Family Medicine, Medical University of Silesia, Zabrze, Poland
  4. 4
    Silesian Analytical Laboratories, Katowice, Poland
  5. 5
    Steering Committee Lipidogram 2004 Study, Warsaw, Poland
  6. 6
    Department of Internal Diseases, Nephrology and Diabetology, Medical University of Silesia, Zabrze, Poland
  7. 7
    Department of Family Medicine, Medical College Jagiellonian University, Krakow, Poland
  8. 8
    Department of Coronary Disease, Medical College Jagiellonian University, Krakow, Poland
  9. 9
    Department of Hypertension, Vascular Diseases and Internal Diseases, University of Medical Sciences, Poznan, Poland
  1. Correspondence to Dr Maciej Tomaszewski, Department of Cardiovascular Sciences, University of Leicester, Clinical Sciences Wing, Glenfield General Hospital, Leicester LE3 9QP, UK; mt142{at}


Background: Excessive body weight is known to cluster with cardiovascular (CV) risk factors, but it is not clear which anthropometric obesity measure provides best independent predictive value of coronary artery disease (CAD).

Methods and results: We explored associations between CAD and four different obesity measures (body mass index (BMI), waist circumference, waist/height and waist/height2) in a cohort of 16 657 subjects (40.4% men; 20.8% CAD patients), recruited by 700 primary care physicians in 444 Polish cities. 42.8% of subjects were classified as overweight, 31.7% as obese and 39.8% had abdominal obesity. In univariate analyses all obesity measures correlated with CAD (p>0.001), but waist/height2 was the strongest discriminator between CAD patients and controls. Age-adjusted and sex-adjusted analyses confirmed a graded increase in CAD risk across distributions of all four obesity measures—1 standard deviation (SD) increase in BMI, waist, waist/height and waist/height2 increased the odds of CAD by 1.23, 1.24, 1.26 and 1.27, respectively (all p<0.001). In models fully adjusted for CV risk factors, waist/height2 remained the strongest obesity correlate of CAD, being the only independent associate of CAD in men. In a fully adjusted BMI—waist circumference stratified model, sarcopenic obesity (waist > median, BMI < median) and simple obesity (waist and BMI > median) were the strongest independent associates of CAD in men (p = 0.008) and women (p>0.001), respectively.

Conclusion: This cross-sectional study showed that waist/height2 may potentially offer a slightly higher predictive value of CAD than BMI or waist circumference and revealed an apparent sexual dimorphism in correlations between obesity measures and CAD.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • ▸ Additional data are published online only at

  • Funding BK’s research fellowship in the Department of Cardiovascular Sciences, University of Leicester, was supported by the Cardiogenics project of the European Union (LSHM-CT-2006-037593). MT is supported by the British Heart Foundation project grant (PG/06/097). NJS holds a British Heart Foundation Chair of Cardiology.

  • Funding Recruitment and phenotyping in Lipidogram 2004 study was supported by a research grant from Schwarz Pharma, Silesian Analytical Laboratories (the executive medical faculty of the study) and the Polish College of Family Physicians. Biochemical analyses were conducted in the executive medical facility of the study, the Silesian Analytical Laboratories.

  • Competing interests None declared.

  • Provenance and Peer review Not commissioned; externally peer reviewed.