Predictors of stroke pattern in hypertensive patients

Background and purpose: Hypertension is a recognized risk factor for stroke. However, it is not clear why hypertensive patients may have different types and causes of stroke. Methods: The possible role of coexisting factors was studied in 1057 patients with hypertension and first stroke admitted to a population-based stroke center. We used logistic regression analysis (multivariate and polychotomous) and the Lausanne Stroke Registry definitions of cerebral hemorrhage, cerebral infarction, lacunar infarction, cardioembolism and large artery disease. Results: Cerebral hemorrhage was not more frequent in these hypertensive patients (111/1057 = 10.5%) than in the total population of the registry (213/2145 = 10%). Cerebral infarction was associated with age above the median (69 years), diabetes, smoking history: family history of cardiac or cerebrovascular disease, hypercholesterolemia and previous TIA. In the ischemic group (n = 946), patients suffered from a lacunar infarction in 36% of the cases, large artery disease was present in 21% of them and cardioembolism in 14.5%. Large artery disease correlated with smoking history (odds ratio: 2.14; 95% confidence intervals: 1.55-2.95) and previous TIA (OR:1.96; 95% CI:1.39-2.75). Cardioembolism con-elated with age over 69 years (OR:2.13; 95% CI:1.47-3.09), In patients with previous TIA and smoking history, large artery disease was presumed in 45% of the cases (vs 21%). Patients aged over 69, suffered a cardioembolic stroke in 29% of the cases (vs 14.5%). Lacunar infarction occurred in 49% of patients aged less than 70 years when no history of smoking or previous TIA was present (vs 36%). Conclusion: The associations between hypertension, TIA, age and smoking history may partially explain why patients with hypertension are more likely to develop a cerebral hemorrhage, lacunar infarction, cardioembolic or a large artery disease stroke.