Association between periodontitis and hemorrhagic stroke among Koreans: a case-control study.

BACKGROUND Several studies suggested that periodontitis is a risk factor for stroke, but the relationship between periodontitis and hemorrhagic stroke has not been widely reported. This study aims to evaluate the association between periodontitis and hemorrhagic stroke and to identify the risk group for this association. METHODS We recruited 165 patients who were diagnosed via computed tomography brain imaging as having had a hemorrhagic stroke and 214 non-stroke control subjects for a case-control study. All participants underwent a clinical periodontal examination using clinical attachment level (CAL) as a marker. Information about sociodemographic factors, behavioral factors, systemic health, and a familial history of systemic health was gathered through an interview using structured questionnaires. The association between periodontitis and hemorrhagic stroke was evaluated using multivariate logistic regression analyses with adjustment for age, gender, income, education, hypertension, diabetes, body mass index, cardiac disease, familial hypertension history, familial diabetes history, familial cardiac disease history, smoking, and alcohol consumption. Subgroup analyses were also performed to investigate potential risk groups. RESULTS After controlling for potential confounders, periodontitis (CAL > or =6 mm) was found to be significantly associated with hemorrhagic stroke (odds ratio: 2.5; 95% confidence interval: 1.1 to 5.6), but this association did not exhibit a dose-dependent response for periodontitis (percentile of sites of periodontal pockets with CAL > or =5 mm among total probed pockets). The association between periodontitis (CAL > or =6 mm) and hemorrhagic stroke was significant for males, patients who had a lower income than control subjects, obese patients, and patients without diabetes. CONCLUSIONS Periodontitis may be an independent risk factor for hemorrhagic stroke. Risk groups include males, patients without diabetes, and obese subjects.