Intended for healthcare professionals


Association of polymorphism of human α oestrogen receptor gene with coronary artery disease in men: a necropsy study

BMJ 2000; 321 doi: (Published 29 July 2000) Cite this as: BMJ 2000;321:273
  1. Tarja A Kunnas (tarja.kunnas{at}, research fellowa,
  2. Pekka Laippala, professorb,
  3. Antti Penttilä, professorc,
  4. Terho Lehtimäki, senior research fellowd,
  5. Pekka J Karhunen, professora
  1. a Department of Forensic Medicine, Medical School, FIN-33014 University of Tampere, Tampere University Hospital, Tampere, Finland
  2. b School of Public Health, FIN-33014 University of Tampere
  3. c Department of Forensic Medicine, PO Box 40, FIN-00014 University of Helsinki, Helsinki, Finland
  4. d Laboratory of Atherosclerosis Genetics, Department of Clinical Chemistry, FIN-33014 University of Tampere
  1. Correspondence to: T A Kunnas
  • Accepted 14 April 2000

The higher oestrogen concentrations in women have been suggested as the reason for their slower development of atherosclerosis compared with men. Oestrogen receptors have been located on macrophages, smooth muscle cells, and endothelial cells in women and men, but it is not known whether the protective effect at the level of the arterial wall is mediated by these receptors. It has been reported that premenopausal women have fewer α oestrogen receptors in atherosclerotic than in normal coronary arteries.1 The gene for human α oestrogen receptor contains a polymorphism in the regulatory (upstream) region of the gene: this polymorphism consists of a dinucleotide (thymine and adenine) repeat, the length of which has been associated with bone mineral density, suggesting an effect on oestrogen receptor transcription.2 This prompted us to study whether this polymorphism is associated with coronary artery disease in men.

Subjects, methods, and results

The associations of the polymorphism with atherosclerosis and myocardial infarction were studied in the Helsinki sudden death study, a prospective series of necropsies of white Finnish men who died suddenly.3 Atherosclerotic changes in the coronary arteries were measured by computer assisted planimetry, and coronary narrowings were determined from plastic casts.3 The presence of myocardial infarction was confirmed by macroscopic and histological examination of the myocardium. We selected the 119 cases (with mean age 53.4 (SD 8) years) for our analyses according to phenotype—52 men with severe coronary atherosclerosis (mean coronary stenosis 65.9% (10.8%)) and 67 men with only slightly narrowed coronary arteries (mean 22.1% (13.7%)). The causes of deaths were coronary heart disease in 59, violent death or accident in 38, and other diseases in 22. DNA was extracted from blood by a standard method, amplified by polymerase chain reaction, and analysed by capillary gel electrophoresis as described previously.4

Because of the large number of dinucleotide repeats,4 we used the median number of the repeat (n=19) to categorise the study population into three groups: those with short allele genotypes (both alleles of <19 repeats), those with long allele genotypes (both alleles of ≥19 repeats), and those with mixed genotypes (one short and one long allele). In analysis of covariance, with age and body mass index as covariates, men with long allele genotypes had a significantly greater number of severely narrowed coronary arteries (P=0.009), larger areas of complicated lesions (P=0.008), and more calcification of the coronary arteries (P=0.01) than men with short alleles (see figure). Atherosclerotic changes in the mixed genotype group were intermediate and not significantly different from those in the short or long allele genotype group.


A stepwise logistic regression analysis with age and body mass index as covariates showed that the group with long allele genotypes had a higher risk of myocardial infarction compared with the group with short allele genotypes (odds ratio 4.4 (95% confidence interval 1.21 to 15.70); P=0.025) and seemed to be more predisposed to coronary thrombosis (odds ratio 11.4 (1.2 to 108.8); P=0.04). When all confounding factors (age, body mass index, smoking, alcohol consumption, diabetes, hypertension) were forced in statistical models, the tendencies of the results were the same, but the small number of cases with all data (n=42) weakened the statistical significance.


Our preliminary results suggest that the length of the dinucleotide repeat in the regulatory region of the α oestrogen receptor gene is associated with severity of coronary artery disease in men. This may partly explain differences between individuals in the development of coronary artery disease. Of the three other polymorphisms in this gene, the most extensively studied, intronic PvuII polymorphism, showed no association with coronary artery stenosis as measured by angiography.5 Although the biochemical evidence is presently lacking, we speculate that carriers of the long repeat variants have lower expression of the oestrogen receptor gene and benefit less from the cardiovascular protective effect of oestrogen receptors.


We thank Outi Lumme and Mervi Niittylahti for their skilful technical assistance, Seppo Tyynelä for planimetric measurements, and Markus Perola for DNA isolation.

Contributors: TAK carried out the genotyping, did most statistical analyses, and wrote the first draft of the article. PL was responsible for statistical analyses. AP was involved in sample and data collection. TL participated in writing the article. PJK initiated the study, was involved in sample and data collection, and is the guarantor for the study. All authors helped in completing the article.


  • Funding Funding: This work was funded by the Medical Research Fund of Tampere University Hospital, the Yrjö Jahnsson Foundation, the Finnish Foundation of Alcohol Research, and the Tampere Regional Fund of the Finnish Cultural Foundation.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract