Background and Purpose: High sensitivity C-reactive protein (hsCRP), soluble intercellular adhesion molecule-1 (sICAM-1) and fibrinogen are markers of systemic inflammation that have been associated with a greater risk and severity of total and ischemic stroke, as well as with elevated blood pressure, a primary risk factor for stroke. However, few studies have examined whether the role of these inflammatory risk factors on stroke pathophysiology is influenced by hypertension status.
Methods: Blood samples were collected and assayed for hsCRP, sICAM-1, and fibrinogen among 27,330 initially healthy women from the Women’s Health Study who were followed from 1992-2014. Self-reported hypertension status, cardiovascular risk factors, lifestyle and alcohol consumption were obtained from the baseline questionnaire prior to randomization. New cases of stroke were ascertained by self-report on annual follow-up questionnaires and counted only if confirmed by medical records according to the National Survey of Stroke criteria. The primary outcome was total stroke with ischemic and hemorrhagic evaluated as secondary outcomes. Multivariable Cox models were utilized to estimate the overall association between each marker of inflammation and stroke, and also separately stratified by hypertension status.
Results: We observed 629 incident total strokes over 477, 278 person-years. In overall multivariable analyses, the highest quartile of hsCRP and sICAM-1 were associated with a significantly greater risk of total stroke compared to the lowest quartile (hsCRP: HR=1.76, 95% CI: 1.38, 2.24, p-trend<0.001; sICAM-1: HR=1.29, 95% CI: 1.01, 1.65, p-trend =0.01). Fibrinogen was not significantly associated with risk of total stroke in multivariable models. All main effect estimates were attenuated when adjusted for baseline hypertension status. In analyses stratified by hypertension status at baseline, elevated hsCRP (extreme quartiles) was significantly associated with risk of total stroke only among prehypertensive (p-trend=0.03) and hypertensive women (p-trend=0.07), although the interaction was not statistically significant. In analyses by stroke type, the findings for ischemic stroke were similar to those of total stroke, while there was no significant association observed with hemorrhagic stroke for any of the measured markers of inflammation, among the limited number of events (n=111).
Conclusions: In this cohort of women free of cardiovascular disease at baseline, elevated hsCRP and sICAM-1 were significantly associated with risk of stroke. Furthermore, the association between hsCRP and stroke may be restricted to prehypertensive and hypertensive women.