Oxidative Damage in Ischemic Stroke Revealed Using Multiple Biomarkers
Background and Purpose—We investigated changes in oxidative damage after ischemic stroke using multiple biomarkers.
Methods—Serial blood and urine samples of ischemic stroke subjects and age-matched control subjects were assayed for F2-isoprostanes, hydroxyeicosatetraenoic acid products, F4-neuroprostanes, 24-hydroxycholesterol, allantoin, and urate.
Results—Sixty-six stroke subjects (mean age, 65 years; median National Institutes of Health Stroke Scale 17) and 132 control subjects were recruited. A bimodal pattern of change was observed in plasma and urinary F2-isoprostanes and plasma 24-hydroxycholesterol. The rise in plasma hydroxyeicosatetraenoic acid products, F4-neuroprostanes, and allantoin was highest 6 to 12 hours after stroke onset, whereas plasma urate was significantly lower than controls on Days 1 to 3. After adjusting for age and baseline National Institutes of Health Stroke Scale, baseline plasma esterified hydroxyeicosatetraenoic acid products (OR, 1.01; 95% CI, 1.01 to 1.02), plasma urate (1.01; 1.00 to 1.01), and plasma free F4-neuroprostanes (2.73; 1.76 to 3.93) were associated with 90-day good functional recovery (modified Rankin Scale ≤1).
Conclusions—Multiple markers of oxidative damage are increased immediately after stroke and remain elevated for several days. Recognition of these temporal changes may help design better antioxidant treatment trials for acute ischemic stroke.
Detailed biomarker studies, incorporating multiple time points during the early events and subsequent course of stroke recovery, are necessary to evaluate the contribution of oxidative damage in stroke. Previous studies that examined biomarkers of oxidative damage and antioxidant levels tend to be limited by a short duration of follow-up, lack of suitable control subjects, inadequate time points, and suboptimal choice or analysis of biomarkers (Supplemental Table I; http://stroke.ahajournals.org). We prospectively investigated changes in multiple markers of oxidative damage after ischemic stroke as compared with age-matched control subjects.
Materials and Methods
Consecutive patients with ischemic stroke who presented within 6 hours of symptom onset to the National University Hospital, Singapore, were included. Information on demographics, vascular risk factors, stroke severity and subtype was prospectively collected. Control subjects closely matched the characteristics of the stroke cohort in a ratio of 2 controls for each stroke subject. Power calculations, performed a priori on the primary variables from our pilot study,1 indicated that a minimum sample size of 60 was required to assess differences in markers of oxidative damage between stroke subjects and control subjects. The study protocol was approved by the Domain Specific Review Board, National Healthcare Group, Singapore, and all patients (or immediate next of kin) provided written informed consent.
Peripheral blood and urine samples, obtained immediately on presentation (before thrombolytic treatment) and serially 6 hours, 12 hours, 18 hours, and Days 1 to 4 and 7 and 14 after stroke onset, were assayed for F2-isoprostanes, hydroxyeicosatetraenoic acid products (HETEs), F4-neuroprostanes, 24-hydroxycholesterol, allantoin (a product of oxidative damage to urate), urate, phospholipase A2, and platelet-activating factor acetylhydrolase activities. The threshold for significance was adjusted for multiple comparisons (with the Bonferroni method).
Sixty-six patients with ischemic stroke and 132 age-matched control subjects were recruited; their characteristics are summarized in Supplemental Table II. Fifty (76%) patients received intravenous thrombolysis and median collection time for initial blood and urine samples was 171 minutes (range, 122 to 245 minutes) from stroke onset. Plasma F2-isoprostanes, HETEs, F4-neuroprostanes, 24-hydroxycholesterol, and allantoin levels were significantly increased in these initial samples (Table). Biomarker levels did not differ between stroke subtypes and the recanalization status of the occluded artery (data not shown). Inflammatory status (measured using high-sensitivity C-reactive protein) did not differ between atrial fibrillation and nonatrial fibrillation patients.
Isoprostanes are the best available biomarker of lipid peroxidation. Plasma esterified F2-isoprostanes tended to be higher than controls at all time points and this was significant immediately and on Day 3 after stroke onset, whereas plasma free F2-isoprostanes were elevated immediately and up to Day 7 (Figure 1A–B). Free F2-isoprostanes can be oxidized to 2,3-dinor-F2-isoprostanes, which are subsequently reduced to 2,3-dinor-5,6-dihydro-F2-isoprostanes, all of which are excreted in urine. A similar bimodal pattern of change was observed in urinary F2-isoprostanes, the levels appearing to peak immediately and 2 to 3 days after stroke onset. Urinary levels of 8-iso- and 2,3-dinor-F2-isoprostanes were elevated, whereas the levels of urinary 2,3-dinor-5,6-dihydro-F2-isoprostanes were decreased in stroke subjects (Figure 2A–D).
The esterified and free forms of plasma HETEs (an inflammatory biomarker) were significantly higher immediately and during the subsequent stages of stroke, the rise being highest 6 to 12 hours after stroke onset (Figure 1C–D). The levels of urinary HETEs were significantly higher immediately and on Days 1 to 2 after the onset of stroke (Figure 2D).
The levels of esterified and free forms of plasma F4-neuroprostanes were significantly higher immediately and during the subsequent stages of stroke as compared with controls. The rise in plasma F4-neuroprostanes was highest approximately 12 hours after the onset of stroke (Supplemental Figure IA–B).
Plasma 24-hydroxycholesterol was significantly higher immediately and on Days 1 to 7 after stroke onset (Supplemental Figure IIA). Plasma allantoin was significantly higher immediately and 6 hours after the onset of stroke (Supplemental Figure IIB), whereas urate was significantly lower on Days 1 to 3 (Supplemental Figure IIC).
Using logistic regression analyses (adjusting for age and baseline National Institutes of Health Stroke Scale), baseline plasma esterified HETEs (OR, 1.01; 95% CI, 1.01 to 1.02), plasma urate (1.01; 1.00 to 1.01), and plasma free F4-neuroprostanes (2.73; 1.76 to 3.93) were associated with good functional recovery (modified Rankin Scale ≤l).
Multiple reliable markers of oxidative damage were observed to be increased immediately after ischemic stroke and the changes vary according to stroke duration. Recognition of these temporal changes may help choose the optimal timing and duration of antioxidant treatment trials for acute stroke management. This study also provides human data that support the merits of investigating lipid oxidation and related products such as F2-isoprostanes, HETEs, and docosahexaenoate in stroke. Our data suggest that urate and F4-neuroprostanes could be beneficial factors in stroke outcome.
Sources of Funding
We acknowledge support from the National Medical Research Council, Singapore (Grant NMRC/1157/2008).
The online-only Data Supplement is available at http://stroke.ahajournals.org/cgi/content/full/STROKEAHA.110.618835/DC1.
- Received February 24, 2011.
- Accepted March 22, 2011.
- © 2011 American Heart Association, Inc.