|
|
||||||||||
|
J Am Coll Cardiol, 2003; 42:661-667, doi:10.1016/S0735-1097(03)00781-2 © 2003 by the American College of Cardiology Foundation |



,*
* Department of Clinical Biochemistry, Herlev University Hospital, Herlev, Denmark
The Copenhagen City Heart Study, Bispebjerg University Hospital, Copenhagen, Denmark
Department of Clinical Biochemistry, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
Manuscript received January 28, 2003; revised manuscript received April 25, 2003, accepted May 9, 2003.
* Reprint requests and correspondence: Dr. Børge G. Nordestgaard, Department of Clinical Biochemistry, Herlev University Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark.
brno{at}herlevhosp.kbhamt.dk
| Abstract |
|---|
|
|
|---|
BACKGROUND: The GP IIb/IIIa PlA1/PlA2 polymorphism influences aggregation of platelets; however, an association between ischemic cardiovascular disease and heterozygosity remains controversial, and association with homozygosity is largely unexplored.
METHODS: We genotyped the participants of the Copenhagen City Heart Study, a prospective cardiovascular investigation of the Danish general population (n = 9,149, 22-year follow-up) and assessed the risk of ischemic cardiovascular disease in heterozygotes or homozygotes versus noncarriers.
RESULTS: Of the participants, 70.0%, 27.3%, and 2.7% were noncarriers, heterozygotes, or homozygotes, respectively. Incidence of ischemic cardiovascular disease was 167 and 103 per 10,000 person-years in homozygous and noncarrier men (log-rank: p = 0.006), whereas this difference was not observed in women (p = 0.33) (genotype·gender interaction: p = 0.03). In homozygous versus noncarrier men <40 years of age, 40 to 50 years, and >50 years at entry, age-adjusted relative risks (RRs) of ischemic cardiovascular disease were 3.6 (1.4 to 9.0), 2.4 (1.3 to 4.6), and 1.0 (0.6 to 1.8), respectively (age·genotype interaction in men: p = 0.04); equivalent multifactorially adjusted RRs were 3.0 (1.1 to 8.0), 2.0 (1.0 to 3.9), and 1.0 (0.6 to 1.8), respectively. The corresponding age-adjusted RR values of MI in men were 5.2 (1.5 to 18), 3.5 (1.6 to 7.5), and 0.5 (0.1 to 1.5), respectively (age·genotype interaction in men: p = 0.002); equivalent multifactorially adjusted RRs were 3.8 (1.0 to 15), 3.1 (1.4 to 6.9), and 0.5 (0.2 to 1.5), respectively.
CONCLUSIONS: PlA2/PlA2 homozygosity is associated with a three-fold and four-fold risk of ischemic cardiovascular disease and MI in young men.
| ||||||||||||
In a recent meta-analysis (4) of mainly case-control studies with conflicting results encompassing more than 17,000 individuals, aggregated results showed an odds ratio for ischemic cardiovascular disease in PlA1/PlA2 heterozygotes combined with PlA2/PlA2 homozygotes versus PlA1/PlA1 noncarriers of 1.10 (95% confidence interval [CI] 1.03 to 1.18). Only two prospective studies have been published so far (5,6), and all studies published have been too small to specifically investigate homozygosity, let alone having been stratified for age and gender.
We tested the hypothesis that platelet GP IIb/IIIa homozygotes or heterozygotes versus noncarriers have increased risk of ischemic cardiovascular disease and myocardial infarction (MI), overall or stratified for age and gender. For this purpose we used the Copenhagen City Heart Study, a prospective cardiovascular study of the Danish general population with 9,149 participants and a total of 22 years' follow-up.
| Methods |
|---|
|
|
|---|
Myocardial infarction was classified according to World Health Organization International Classification of Diseases 8th ed. (ICD-8) code 410, or 10th ed. (ICD-10) codes I21-I22. Non-MI ischemic heart disease was ICD-8 codes 411-413 or ICD-10 codes I20 and I25: diagnosis was based on characteristic symptoms of stable angina pectoris according to the guidelines of the European Society of Cardiology (location, character, and duration of pain and the relation of pain to exercise [14]).
Ischemic stroke was ICD-8 codes 432-434 or ICD-10 code I63, excluding computed tomography proven hematoma, subarachnoidal hemorrhage, or transient ischemic attack. Individual diagnoses were verified by experienced cardiologists and neurologists, respectively (15). Participants with disease before study entry were excluded from the study (n = 93).
We had 99.97% follow-up; three individuals were lost. More than 99% were whites of Danish descent. All participants gave written informed consent. The ethical committees of Copenhagen and Frederiksberg approved the study (No. 100.2039/91).
The PlA2-allele, also called HPA-1b or Zw(b), is caused by a T
C substitution in nucleotide 176 after A in the start-codon in the gene integrin ß3 (GenBank AccNo NM_000212). This polymorphism was examined as earlier described (16): in short, a 268-bp polymerase chain reaction-fragment covering the whole of exon 3 (GenBank AccNo M32672) was amplified from genomic deoxyribonucleic acid by the use of intronic primers: sense: TTCTGATTGCTGGACTTCTCTT; antisense: TCTCTCCCCACGGCAAAGAGT. After thermocycling, the polymerase chain reaction was digested with the restriction endonuclease MspI, run on a 3% agarose gel and visualized using ethidium bromide. The PlA1/PlA1 noncarrier genotype produced a 221/38/8-bp-pattern, PlA1/PlA2 heterozygous genotype a 221/177/44/38/8-bp-pattern, and PlA2/PlA2 homozygous genotype a 177/44/38/8-bp-pattern.
Blood samples were drawn in the nonfasting state. Colorimetric and turbidimetric assays were used to measure plasma levels of total cholesterol, high-density lipoprotein cholesterol, triglycerides, fibrinogen (all Boehringer Mannheim, Mannheim, Germany), and lipoprotein(a) total mass (DAKO A/S, Glostrup, Denmark). The diagnosis of diabetes mellitus was assigned according to participants' own knowledge of their disease status. The status of smoker was assigned to current and former smokers.
The statistical software package SPSS (SPSS for Windows, release 10.0.7, SPSS Inc., Chicago, Illinois) was used. A p value <0.05 on a two-sided test was considered significant. For characteristics of participants, we used the Student t test on untransformed or log-transformed variables, Mann-Whitney U test, or the Pearson chi-square test. Multifactorial adjustment included all other parameters listed in Table 1 in an analysis of covariance or logistic regression model. Continuous variables were divided in gender-specific tertiles. It was decided a priori to stratify main analyses for gender and age. We plotted cumulative disease incidence against follow-up time using the Kaplan-Meier method, and we tested differences between genotypes using log-rank statistics. Visual inspection of log-minus-log curves was employed to exclude disproportion of hazards over time. Relative risk for disease with 95% CI was calculated using the Cox regression, unadjusted, adjusted for age at entry in decades, or all the factors listed in Table 1 (multifactorial adjustment). We tested for multiplicative interactions among gender, age, and genotype-status in predicting disease, by introducing two-factor interaction terms after inclusion of individual parent terms in the Cox regression model.
|
| Results |
|---|
|
|
|---|
|
|
|
Age at first MI was 61 ± 2.5 years (mean ± SE) and 66 ± 0.5 years in homozygous and noncarrier men (Table 4; Mann Whitney U test: p = 0.03); equivalent values for ischemic cardiovascular disease were 64 ± 2.0 and 67 ± 0.5 years (p = 0.07). This difference was not observed between homozygotes and noncarrier women, or between heterozygotes and noncarriers of either gender (Table 4) .
|
| Discussion |
|---|
|
|
|---|
The more than 30 case-control studies published about this polymorphism and risk of ischemic cardiovascular disease reach divergent results (3,4). Two factors complicate a direct comparison of our results with the findings of others. Most importantly, the majority of previous studies combine PlA1/PlA2 heterozygotes and PlA2/PlA2 homozygotes versus PlA1/PlA1 noncarriers in their calculations in order to obtain sufficient statistical power, and thus do not study homozygotes alone. Second, most previous studies did not investigate associations stratified on gender or age, which proved essential in our work. Our study suggests an association between homozygosity and ischemic cardiovascular disease in men, but not in women. This association is most pronounced in men <40 years of age, an age group where very few women are affected. Among the group 40 to 50 years of age, we do find cases among women and also an elevated relative risk for ischemic cardiovascular disease in homozygous men, but not in homozygous women, thus supporting the gender-specific association.
Contrary to most other reports, we do not find any increased risk of ischemic cardiovascular disease among PlA1/PlA2 heterozygotes versus PlA1/PlA1 noncarriers. Several factors could contribute to this discrepancy: publication bias toward positive results in small studies (4), different ethnicity between studies, differences in study design, and chance alone. Our study is the largest so far and, together with the two other published prospective studies (5,6), agree that PlA1/PlA2 heterozygosity does not increase the risk of ischemic cardiovascular disease.
Of the many in vitro studies examining the impact of the PlA2-allele on platelet function, two (19,20) investigated all three genotypes separately. In these studies, PlA2 was associated with increased platelet aggregability, and thus the propensity to initiate thrombus formation, in a gene-dose dependent manner. Biologically, the observation of increased risk of ischemic cardiovascular disease in PlA2/PlA2 homozygotes therefore seems reasonable. However, this remains an untested correlation that may or may not explain the enhanced risk of homozygotes reported in our study.
Interestingly, when exposed to physiologic concentrations of estrogen, the aggregability of PlA1/PlA2 heterozygous platelets was inhibited more than PlA1/PlA1 noncarrier platelets (21). This accords with our data, which demonstrate an association between PlA2/PlA2 homozygosity and ischemic cardiovascular disease in men only.
Prothrombotic conditions like increased platelet aggregability is believed by some investigators (22,23) to outweigh the importance of atherosclerosis in premature MI, mainly because of their findings of fewer stenoses on coronary angiography in young patients when compared with old patients with MI. Therefore, having the above-mentioned in vitro studies on PlA-genotype and platelet aggregability in mind (1921,24), our finding of age-dependency in men of the effect of the PlA2/PlA2-genotype on MI could simply be caused by increased platelet aggregability induced by the PlA2/PlA2-genotype operating at an early age.
In a series of autopsy studies on men, Mikkelsson et al. (2528) reported lower incidence of atherosclerotic lesions but higher incidence of coronary thrombosis among PlA2-carriers than among PlA1/PlA1 noncarriers for participants <60 years, but not for those >60 years. Although the stratification on genotype and age in these studies (age under or above 60 years, genotype noncarriers vs. heterozygotes, and homozygotes combined) differs from our work, those studies nevertheless support our finding of higher risk of disease among younger homozygous men. Unfortunately, no data on the extent of atherosclerosis (including results from coronary angiographies) of the participants of our study are available. Therefore, we do not know whether the young homozygous men in our study have more or less atherosclerosis than expected.
We cannot totally exclude that our findings represent chance findings simply because no other studies have demonstrated this association in young men. However, the association was found in those <40 years old and in 40- to 50-year-old men, and the association decreased stepwise with increasing age.
Another potential limitation of our study is that participants underwent genotyping only if they attended the 1991 to 1994 third examination. The absence of a correlation between PlA2/PlA2 homozygosity and ischemic heart disease in older men could be the result of a dropout of homozygous men at an early age because of premature death or disability. However, the stable Hardy-Weinberg equilibria throughout the age groups (Table 3) do not support such a hypothesis. Furthermore, age percentiles for noncarrier and homozygous men display a linear relationship, as would be expected with no selection against homozygotes (data not shown). Thus, we do not consider selection bias against homozygotes very likely, but if this does apply for our study, it would rather result in a conservative estimate concerning association of the PlA2-genotype with ischemic cardiovascular disease, and thus cannot explain our results.
Systematic misclassification of genotypes in this study is unlikely, because of agreement with the Hardy-Weinberg equilibrium and because the PCR assay included a restriction enzyme site in each person assayed. Misclassification of ischemic cardiovascular disease in the Copenhagen City Heart Study is very low, as we have 99.97% follow-up, and because all hospital admissions as well as deaths are registered systematically throughout Denmark.
Medication given to prevent ischemic cardiovascular disease and MI might overcome an underlying genetic risk. Therefore, another limitation of the present study is that we do not have complete information on all preventive medication given to all participants during the entire 22-year follow-up; however, addition of the use of antihypertensive and cholesterol-lowering medications at the 1991 to 1994 examination to the multifactorially adjusted models only resulted in trivial changes in the RR estimates.
Finally, the main finding of our prospective study in the Copenhagen City Heart Study, with a total follow-up of 163,987 person-years, is the three-fold and four-fold 22-year risk of ischemic cardiovascular disease and MI in men <40 years homozygous for platelet GP IIb/IIIa PlA2/PlA2.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
32 women are protected while
43 and
44 men are susceptible to ischemic heart disease. The Copenhagen City Heart Study. J Am Coll Cardiol. 2000;35:11921199This article has been cited by other articles:
![]() |
J. Mikkelsson, M. Perola, and P. J. Karhunen Genetics of Platelet Glycoprotein Receptors: Risk of Thrombotic Events and Pharmacogenetic Implications Clinical and Applied Thrombosis/Hemostasis, April 1, 2005; 11(2): 113 - 125. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |