BACKGROUND AND PURPOSE: Intracerebral hemorrhage constitutes an often fatal sequela of thrombolytic therapy in patients with ischemic stroke. Early blood-brain barrier disruption may play an important role, and the astroglial protein S100B is known to indicate blood-brain barrier dysfunction. We investigated whether elevated pretreatment serum S100B levels predict hemorrhagic transformation (HT) in thrombolyzed patients with stroke. METHODS: We retrospectively included 275 patients with ischemic stroke (mean age of 69+/-13 years; 46% female) who had received thrombolytic therapy within 6 hours of symptom onset. S100B levels were determined from pretreatment blood samples. Follow-up brain scans were obtained 24 hours after admission, and HT was classified as either hemorrhagic infarction (1, 2) or parenchymal hemorrhage (1, 2). RESULTS: HT occurred in 80 patients (29%; 45 hemorrhagic infarction, 35 parenchymal hemorrhage). Median S100B values were significantly higher in patients with HT (0.14 versus 0.11 mug/L; P=0.017). An S100B value in the highest quintile corresponded to an OR for any HT of 2.87 (95% CI: 1.55 to 5.32; P=0.001) in univariate analysis and of 2.80 (1.40 to 5.62; P=0.004) after adjustment for age, sex, symptom severity, timespan from symptom onset to hospital admission, vascular risk factors, and storage time of serum probes. A pretreatment S100B value above 0.23 mug/L had only a moderate sensitivity (0.46) and specificity (0.82) for predicting severe parenchymal bleeding (parenchymal hemorrhage 2). CONCLUSIONS: Elevated S100B serum levels before thrombolytic therapy constitute an independent risk factor for HT in patients with acute stroke. Unfortunately, the diagnostic accuracy of S100B is too low for it to function in this context as a reliable biomarker in clinical practice.
BACKGROUND AND PURPOSE:Intracerebral hemorrhage constitutes an often fatal sequela of thrombolytic therapy in patients with ischemic stroke. Early blood-brain barrier disruption may play an important role, and the astroglial protein S100B is known to indicate blood-brain barrier dysfunction. We investigated whether elevated pretreatment serum S100B levels predict hemorrhagic transformation (HT) in thrombolyzed patients with stroke. METHODS: We retrospectively included 275 patients with ischemic stroke (mean age of 69+/-13 years; 46% female) who had received thrombolytic therapy within 6 hours of symptom onset. S100B levels were determined from pretreatment blood samples. Follow-up brain scans were obtained 24 hours after admission, and HT was classified as either hemorrhagic infarction (1, 2) or parenchymal hemorrhage (1, 2). RESULTS:HT occurred in 80 patients (29%; 45 hemorrhagic infarction, 35 parenchymal hemorrhage). Median S100B values were significantly higher in patients with HT (0.14 versus 0.11 mug/L; P=0.017). An S100B value in the highest quintile corresponded to an OR for any HT of 2.87 (95% CI: 1.55 to 5.32; P=0.001) in univariate analysis and of 2.80 (1.40 to 5.62; P=0.004) after adjustment for age, sex, symptom severity, timespan from symptom onset to hospital admission, vascular risk factors, and storage time of serum probes. A pretreatment S100B value above 0.23 mug/L had only a moderate sensitivity (0.46) and specificity (0.82) for predicting severe parenchymal bleeding (parenchymal hemorrhage 2). CONCLUSIONS: Elevated S100B serum levels before thrombolytic therapy constitute an independent risk factor for HT in patients with acute stroke. Unfortunately, the diagnostic accuracy of S100B is too low for it to function in this context as a reliable biomarker in clinical practice.
Authors: Zhiqun Zhang; Stefania Mondello; Firas Kobeissy; Richard Rubenstein; Jackson Streeter; Ronald L Hayes; Kevin K W Wang Journal: Transl Stroke Res Date: 2011-12-07 Impact factor: 6.829