171 Combined inflammatory biomarkers predict recurrent ipsilateral ischaemic stroke in patients with carotid atherosclerosis

Abstract Background Large-artery ischaemic stroke is associated with a high-risk of recurrence. Inflammation is important in the development and rupture of carotid plaque. It remains unknown if blood-biomarkers can reliably quantify vascular inflammation. We investigated whether blood-biomarkers were associated with recurrent ipsilateral Ischaemic Stroke (IS). Methods hsCRP and IL-6 were measured in blood-samples using immunochemiluminescence immunoassays from three cohort studies. Median follow-up was 3.9 years (IQR 1.6–5.8). The primary outcome was recurrent ipsilateral IS. Results Of 269 participants, 130 (48.3%) had severe CS (median age, 71 (IQR 64–78)). There were 19 ipsilateral recurrent IS in 1,046-patient-years follow-up. hsCRP was dichotomised at 2 mg/L and IL-6 at 7 pg/ml (expected-normal value for assay). 187 (69.5%) had hsCRP<2 mg/L and 170 (63.2%) had IL-6 ≥ 7 pg/ml. 146 (54.3%) participants had combined-raised hsCRP≥2 mg/L and IL-6 ≥ 7 pg/ml. On unadjusted Cox analysis, hsCRP≥2 mg/L was associated with increased risk of ipsilateral recurrent IS (HR 9.25, CI 1.24–69.2, p = 0.03) while IL-6 ≥ 7 pg/ml trended towards increased risk (HR 2.36, CI 0.79–7.1, p = 0.1.) However, combined raised-hsCRP/IL-6 was associated with an increased risk of recurrent ipsilateral IS (HR 3.58, CI 1.2–10.7, p = 0.02.) However, after adjustment for age, sex, smoking and time-to-phlebotomy, these associations attenuated. When patients were divided into three risk categories defined by levels of hsCRP and IL-6 (low [both markers low], intermediate [discordant markers], high [both markers raised]0, the rate of recurrent stroke increased across the categories, p = 0.013 for trend. The c-statistic for combined-raised hsCRP/IL-6 was 0.66 (CI 0.57–0.75) indicating prognostication better than chance to predict recurrent-ipsilateral IS. Conclusion Using blood inflammatory proteins in combination may add value in risk prediction in patients with carotid atherosclerosis, although this concept requires independent replication and validation in larger cohorts.