Plaques in the carotid artery and risk of cerebral infarcts in the elderly; The Rotterdam Study
Background Presence of carotid plaques is associated with increased risk of cerebral infarcts. Whether they directly cause a stroke through embolism or merely are indicators of generalised atherosclerosis is not extensively investigated. We studied this by assessing the association between plaques at different carotid segments and risk of subtypes of cerebral infarcts. Methods Our study is based on 5444 subjects from the Rotterdam Study, a population based cohort study. Subjects were free from previous stroke. Presence of plaques was assessed in 3 segments of both carotid arteries at baseline (1990–1993). A total plaque score reflected the number of locations with plaques (range 0–6). Follow up for stroke was complete until December 31 1997. We calculated the risk of severe plaques (score 5–6) compared to absence of plaques (score 0) by Cox regression, adjusting for age and gender. Furthermore, we analysed the risk of unilateral and bilateral plaques compared to absence of plaques at different segments of the carotid artery. Relative risks (95% CI) are given for subtypes of cerebral infarct. Results In total 224 strokes occurred of whom 130 were ischemic. Severe plaques were present in 7% of the population and associated with an increased risk of lacunar and anterior circulation infarcts (RR 12.0 (2.0–70) and 4.5 (1.6–12.3), respectively). After exclusion of infarcts of presumed cardio-embolic cause the relative risk of anterior circulation infarcts was 9.4 (2.5–35.7). The risk for bilateral plaques was higher than for unilateral plaques when we analysed carotid segments separately. We found no association between severe plaques and risk of posterior circulation infarcts. Conclusion Carotid plaques are associated with increased risk of lacunar and anterior circulation infarcts. Associations were stronger with increasing number of plaques. This suggests that carotid plaques are markers of atherosclerosis on large and small vessel level. Since we found no association with posterior circulation infarcts it seems likely that at least part of the effect is through thrombo-embolism.