Sarah Coveney1,2, John J.McCabe1,Murphy Sean2,1,Orina Belton3, M. Crowe4,1, Eamon
Dolan5,1, Tim Cassidy4,1, Monica De Gaetano3, Maria Fitzgibbon6, Joe Harbison2,1,Gillian
Horgan1,Michael Marnane2,1,Aine Merwick7,David Williams1,7, P.J. Kelly2,1
1SCTNI, Dublin, Ireland
2Stroke Service, Mater University Hospital and University College Dublin, Dublin, Ireland
3University College Dublin, Conway Institute, Dublin, Ireland
4St Vincent’s University Hospital,Dublin, Ireland
5James Connolly Memorial Hospital,Dublin, Ireland
6Mater Misercordiae University Hospital,Dublin, Ireland
7Beaumont University Hospital,Neurology,Dublin, and Royal College of Surgeons in Ireland,
Background: Inflammation plays a role in the development of ischaemic cerebrovascular
events. High sensitivity C-reactive protein (CRP) is known to predict recurrent events.
Little data exists for more upstream serum markers of inflammation.
Methods: BIO-STROKE and BIO-TIA were multicentre prospective biomarker and
imaging studies of patients with non-severe stroke, TIA and controls. Exclusion criteria were
malignancy, infection, recent trauma / surgery, recurrent stroke before phlebotomy/MRI.
Serum biomarkers analysed included Interleukin (IL) – 6, CRP, IL-1, IL-8, IL10, IL12p70,
IFN and TNF.Plasma CRP and IL-6 were measured by mass spectrometry. Additional
biomarkers were measured using ELISA. Follow up was performed at 7, 28, 90 days and 1
Results: 680 patients (439 strokes, 241 TIAs) and 68 controls were included in the
analysis. The median age was 70 for cases. Carotid stenosis was present in 23.6% of cases.
Median CRP was 3.75mg/L, 2.36mg/l and 1.87mg/L in the stroke, TIA and control groups
(p=<0.001). Median IL-6 was 5.86pg/ml (stroke), 4.25pg/ml (TIA), 3.06pg/ml (control)
On multivariate cox regression analysis baseline IL6 and CRP were independent predictors
of all cause death at 1 year with a HR of 1.005 (95% CI 1.002-1.007, p<0.001).and
1.005(95% CI 1.002-1.007, p<0.001) per unit increase. Both IL6 and CRP were associated
with vascular death at 1 year. In adjusted analyses, IL6 and CRP were associated with
poor functional outcome at 1 year (OR of 1.02(CI 1.01 -1.03) and 1.02(CI 1.01-1.03) per
unit increase, for IL6 and CRP respectively).
On adjusted analysis, when IL6 was analysed as quartiles, there was a strong association
with death at 1 year with an OR 1.87 (95% CI 1.19-2.93).CRP, analysed as quartiles,
demonstrated an OR for death at 1 year of 1.64 (1.10-2.46).
Conclusion: IL-6 and CRP may be a useful prognostic factor for the prediction of outcome
and death after stroke at 1 year follow up.