Plasma C-Reactive Protein and Clinical Outcomes after Acute Ischemic Stroke: A Prospective Observational Study

Abstract

Background and purpose: Although plasma C-reactive protein (CRP) is elevated in response to inflammation caused by brain infarction, the association of CRP with clinical outcomes after acute ischemic stroke remains uncertain. This study examined whether plasma high-sensitivity CRP (hsCRP) levels at onset were associated with clinical outcomes after acute ischemic stroke independent of conventional risk factors and acute infections after stroke.

Methods: We prospectively included 3653 patients with first-ever ischemic stroke who had been functionally independent and were hospitalized within 24 h of onset. Plasma hsCRP levels were measured on admission and categorized into quartiles. The association between hsCRP levels and clinical outcomes, including neurological improvement, neurological deterioration, and poor functional outcome (modified Rankin scale ≥3 at 3 months), were investigated using a logistic regression analysis.

Results: Higher hsCRP levels were significantly associated with unfavorable outcomes after adjusting for age, sex, baseline National Institutes of Health Stroke Scale score, stroke subtype, conventional risk factors, intravenous thrombolysis and endovascular therapy, and acute infections during hospitalization (multivariate-adjusted odds ratios [95% confidence interval] in the highest quartile versus the lowest quartile as a reference: 0.80 [0.65-0.97] for neurological improvement, 1.72 [1.26-2.34] for neurological deterioration, and 2.03 [1.55-2.67] for a poor functional outcome). These associations were unchanged after excluding patients with infectious diseases occurring during hospitalization, or those with stroke recurrence or death. These trends were similar irrespective of stroke subtypes or baseline stroke severity, but more marked in patients aged <70 years (Pheterogeneity = 0.001).

Conclusions: High plasma hsCRP is independently associated with unfavorable clinical outcomes after acute ischemic stroke.

Fig 1. Subgroup analysis.

Multivariable-adjusted odds ratio and 95% confidence interval of each hsCRP quartile for poor functional outcome at 3 months are shown according to subgroups. Q1–Q4 indicate the four groups according to the quartile of hsCRP (mg/L). The multivariable model included age, sex, baseline NIHSS score, stroke subtypes, hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, smoking, drinking, chronic kidney disease, body mass index, intravenous thrombolytic therapy and endovascular therapy, and acute infections. Subgroups included age (≥70 years or <70 years), stroke subtypes (cardioembolic or others), and stroke severity (baseline NIHSS score 0–4 or NIHSS score ≥5). P for heterogeneity (P_{heterogeneity}) was calculated by means of the interaction term.