Abstract 4067: Antibodies of Bacteria and C-Reactive Protein Confer Risk of Myocardial Infarction
Objectives. We wanted to test the hypothesis that common oral bacteria causing marginal periodontitis are associated with myocardial infarction (MI) and the associated role of the inflammation reactant C-reactive protein.
Methods. The study was based on men born 1923–52 taking part in the Oslo Study in 1972/73 and its follow-up in 2000. A nested case-control study was performed. Cases were men with selfreported history of myocardial infarction (n=548) and age-matched controls (n=625). Serum from blood samples at the screening in 2000 was stored for analyses on antibodies against four oral bacteria: Porphyromonas gingivalis (PG), Actinobacillus actinomycetemcomitans (AA), Tannerella forsythia (TF) and Treponema denticola (TD). The ELISA method was used to quantify the level of antibodies. In addition, the level of C-reactive protein (CRP) was measured.
Results. Chi-square tests of quartile values of optical density levels of antibodies between cases and controls showed a trend (p=0.078) for AA only. Mean level for all four antibodies were higher for cases of MI but the differences were not significant as standard deviations were large. There were significant differences in mean ± SD of CRP of cases 4.28±6.59 versus controls 3.72±8.38; p=0.010. An analytical model was used in logistic regression analyses adjusting for known risk factors for acute myocardial infarction, testing upper quartile value of any one of four antibodies AA, TF, TD or PG versus lower values and showed a significant association to MI (OR=1.31; p=0.039). Equivalently, with any three of the antibodies the association was significant for TD, AA or PD (OR=1.29; p=0.042). Univariate analysis of CRP gave increased risk with increasing quartile value; trend, p=0.016, but not in multivariate analysis including the oral antibodies.
Conclusion. Serum antibodies against major periodontal pathogens were associated with MI but could not be linked to any one of the four studied. CRP was a significant predictor in univariate analysis but was not an independent risk factor for MI in adjusted analyses including the antibodies. Nor did it reduce the risk estimates of the antibodies. These results support the hypothesis that infection due to common oral bacteria is a predictor for myocardial infarction.