Association of cancer and stroke: A population-based study

Abstract

20672 Background: Cancer is considered as a risk factor of stroke. We tried to compare cardiovascular risk factors in patients with cancer and stroke and in patients with isolated stroke. Demographic, clinical and biological data were analyzed. Methods: A data-base was initialized prospectively including all patients hospitalized for stroke in our neurological unit since January, 2004. Patients with intra-tumoral and post-traumatic hemorrhage were excluded. The analysis was realized when 700 patients were included and separated in 2 groups: absence of cancer (group 1), report of cancer (group 2). For every patient, were collected: demographic data, medical history, type, location and mechanism of stroke, cardiovascular risk factors, treatment. When cancer was reported we collected date of diagnosis, location, date and type of reductive treatment (surgery, chemotherapy, radiotherapy, hormonotherapy) and status of neoplasia. Results: On 700 patients (292 women, 408 men, mean age: 65.8 years), 78 (11%) had a history of cancer (group 2). These patients were older (mean age: 70.7 years versus 64.8 years, p=0.036), mostly women (51% versus 41%, p<0.001). No difference was found concerning type and location of stroke. Stroke of unknown origin was more frequent in group 1, lacunar type in group 2. The cardiovascular risk factors were identical in the 2 groups with the exception of type II diabetes (more frequent in the group 1) and biological abnormalities (more frequent in the group 2). In group 2, specific data were collected: aseptic endocarditis (1 case), venous thrombosis with carcinomatous meningitis, intra vascular coagulation, stroke during chemotherapy or hormonotherapy, radiation-induced carotid stenosis, concomitant diagnosis of cancer. Conclusions: The prevalence of the association stroke-cancer was 11% and corresponds to the literature. The simultaneous discovery of a cancer remained exceptional (6/700). Cardiovascular risk factors were similar in the 2 groups. The role played by biological abnormalities and cancer treatments is discussed. Association of cancer and stroke is a frequent phenomenon but the role of cancer is not always obvious. Our study fits with the literature. No significant financial relationships to disclose.