RESUMO
BACKGROUND: Periodontitis increases the risk of atherosclerotic cardiovascular disease and ischemic stroke. In this study, we evaluated whether serum antibody levels against individual periodontal pathogens are significantly associated with ischemic stroke subtypes and their risk factors. METHODS: Patients with acute ischemic stroke (n = 132; 74 male and 58 female, 71.3 ± 10.7 years) and patients with no previous stroke (n = 77; 38 male and 39 female, 70.7 ± 9.5 years) were consecutively enrolled in this study. Stroke subtype was evaluated based on the Trial of Org 10172 in Acute Stroke Treatment classification. Serum was obtained from each patient after obtaining their consent to participate in the study. The levels of serum antibodies against Aggregatibacter actinomycetemcomitans (Aa), Porphyromonas gingivalis (Pg) and Prevotella intermedia (Pi) were evaluated by ELISA. Serum high-sensitivity C-reactive protein (hs-CRP) levels were measured by nephelometry. RESULTS: Serum hs-CRP levels were significantly associated with acute ischemic stroke even after controlling for acute ischemic stroke, hypertension, diabetes mellitus and bulb/ internal carotid artery (ICA) atherosclerosis which were statistically selected (coefficient 0.245, 95% CI 0.142-0.347, p < 0.0001). The serum-antibody level of Pi was significantly higher in atherothrombotic-stroke patients than in patients with no previous stroke (p = 0.0035). Detectable serum anti-Pg antibody was significantly associated with atrial fibrillation (overall χ(2) = 35.5, R(2) = 0.18, n = 209, p < 0.0001; anti-Pg antibody: OR 4.36, 95% CI 1.71-12.10, p = 0.0017), and detectable serum anti-Pi antibody was significantly associated with bulb/ICA atherosclerosis after controlling for the statistically selected associated factors (overall χ(2) = 46.1, R(2) = 0.18, n = 209, p < 0.0001; anti-Pg antibody: OR 16.58, 95% CI 3.96-78.93, p < 0.0001). The levels of serum anti-Pi antibody were significantly associated with atherothrombotic stroke with the statistically selected associated factors excluding bulb/ICA atherosclerosis (overall χ(2) = 77.0, R(2) = 0.44, n = 129, p < 0.0001; anti-Pi antibody: OR 23.6, 95% CI 2.65-298.2, p = 0.008). However, when we included bulb/ICA atherosclerosis in this model, the levels of serum anti-Pi antibody were no longer significantly associated with atherothrombotic stroke (overall χ(2) = 98.0, R(2) = 0.56, n = 129, p < 0.0001; anti-Pi antibody: p = 0.107). CONCLUSIONS: Our results suggest that anti-Pg antibody is associated with atrial fibrillation and that anti-Pi antibody is associated with carotid artery atherosclerosis. In addition, anti-Pi antibody may be associated with atherothrombotic stroke through its association with carotid artery atherosclerosis. Thus, periodontitis may lead to serious systemic diseases.