Multiple biomarkers covering several pathways improve predictive ability for cognitive impairment among ischemic stroke patients with elevated blood pressure.

Background and Aims: We aimed to evaluate the ability of multiple novel biomarkers representing several pathophysiological pathways to improve risk prediction of post-stroke cognitive impairment.
Methods: We conducted a prospective multicenter study in 638 ischemic stroke patients with elevated blood pressure based on a random subsample from China Antihypertensive Trial in Acute Ischemic Stroke and measured 12 circulating biomarkers in these participants. Cognitive impairment was assessed at 3 months after stroke with definitions of Mini-Mental State Examination (MMSE) score <27 or Montreal Cognitive Assessment (MoCA) score <25.
Results: According to MMSE score, 1 SD increase of rheumatoid factor (odds ratio [OR] 1.22, 95% confidence interval [CI] 1.02-1.46), matrix metalloproteinase-9 (OR 1.47, 95% CI 1.22-1.77) and total homocysteine (OR 1.22, 95% CI 1.01-1.49) after log transformation was significantly associated with the risk of post-stroke cognitive impairment. The ORs associated with their simultaneously high levels were 4.89 (95% CI, 2.31-10.35; p _trend <0.001) and 3.09 (95% CI, 1.60-5.98; p _trend <0.001) for cognitive impairment and the severity of cognitive impairment, respectively. Adding these 3 biomarkers to conventional model significantly improved the risk prediction of cognitive impairment (C statistic 0.729 vs. 0.688, p = 0.004; net reclassification improvement = 33.67%, p < 0.001; integrated discrimination index = 4.61%; p < 0.001). Similar significant findings were observed when cognitive impairment was defined by MoCA score.
Conclusions: Combination of rheumatoid factor, matrix metalloproteinase-9 and total homocysteine can improve the risk prediction of cognitive impairment among ischemic stroke patients with elevated blood pressure. Further studies are warranted to validate our findings and explore their roles as potential therapeutic targets.
(Copyright © 2019 Elsevier B.V. All rights reserved.)