| Part One Related Factors of Blood-Brain Barrier Permeability in Acute Ischemic Stroke PatientsAims:Blood-brain barrier(BBB)breakdown is one of pathological changes after acute ischemic stroke(AIS).Increased BBB permeability(BBBP)is believed to be associated with hemorrhagic transformation(HT)and brain edema.We aimed to find out related factors of blood-brain barrier permeability measured by CT perfusion(CTP)in AIS patients.Methods:We retrospectively reviewed our prospectively collected database of AIS patients who performed head CT perfusion and received reperfusion therapy within 9 hours of stroke onset from July 2011 to February 2017.General baseline patient characteristics were collected:age,sex,past history,laboratory tests,National Institutes of Health Stroke Scale(NIHSS),and imaging findings(infarct volume,hypoperfusion volume,leukoaraiosis severity and absent filling of cortical veins).Hypoperfusion regions were determined on CTP Tmax map as region with Tmax>6s by software MI Star.Hypoperfusion region were automatic outlined as regions of interest(ROIs)by MIStar in each slice separately and their mirror ROIs in the contralateral nonischemic hemisphere were created across the midline.Then we superimposed these ROIs on the permeability-surface area product(PS)maps and cerebral blood flow(CBF)maps by each slice,respectively,to calculate mean PS and CBF values within each ROIs.We used relative PS(rPS)to represent BBBP.The rPS in hypoperfusion region(rPShypo-i)and its contralateral mirror regions(rPShypo-c)were calculated based on the following formula:rPS=(PS/CBF)× 100%.Spearman correlation analysis and multiple linear regression were used to assess the relation between stroke severity,patient characteristics,and rPS.Results:A total of 230 patients were included in the analysis.Median age was 73(IQR 62-80)years and 90(39.1%)were female.Median baseline NIHSS score was 13(IQR 8-17)and median onset-to-imaging time was 175(IQR 96-250)min.Median rPShypo-i was 69.79(IQR 37.25-121.68)%,and median rPShyp-c was 1.66(0.97-3.17)%.Females were found to have lower rPShypo-i(B=-0.128,95%CI-0.213--0.043,p=0.003)and rPShypo-c(B=-0.117,95%CI-0.218--0.016,p=0.023).Larger infarct volume was independent positive correlated with rPShypo-i(B=0.002,95%CI 0.001-0.003,p<0.001)and rPShypo-c(B=0.001,95%CI 0.000-0.001,p=0·036).The rPShypo-i also showed independent positive correlations with history of atrial fibrillation(B=0.225,95%CI 0.142-0.308,p<0.001)and baseline NIHSS score(B=0.012,95%CI 0.004-0.019,p=0.004).While,there was a negative correlation between rPShypo-i and percentage of neutrophil(B=-0.003,95%CI-0.006--0.000,p=0.036).Conclusion:BBBP in both the hypoperfusion region and its contralateral mirror region are associated with sex and stroke severity,indicating there might be a diffused BBB damage after AIS.Part Two The prognostic value of Blood-brain Barrier Permeability for Outcomes of Acute Ischemic Stroke patients after Reperfusion therapyAims:We aimed to determine the predictive value of blood-brain barrier permeability(BBBP)measured by computed tomographic perfusion(CTP)for hemorrhagic transformation(HT),brain edema expansion and functional outcome in acute ischemic stroke(AIS)patients who received reperfusion therapy.Methods:We retrospectively reviewed our prospectively collected database of AIS patients who received reperfusion therapy within 9 hours of stroke onset,and performed head CT perfusion before and after treatment,from July 2011 to February 2017.Hypoperfusion region was defined as region with Tmax>6s.Hypoperfusion regions were automatic outlined as regions of interest(ROIs)by MIStar in each slice separately and their mirror ROIs in the contralateral nonischemic hemisphere were created across the midline.Then we superimposed these ROIs on the permeability-surface area product(PS)maps and cerebral blood flow(CBF)maps by each slice,respectively,to calculate mean PS and CBF values within each ROIs.We used relative PS(rPS)to represent BBBP.The rPS in hypoperfusion region(rPShypo-i)and its contralateral mirror regions(rPShypo-c)were calculated based on the following formula:rPS=(PS/CBF)×100%.HT was evaluated on 24-hour follow-up scans according to European Cooperative Acute Stroke Study(ECASS)? criteria.Brain edema was assessed according to Wardlaw's method on baseline and follow-up imaging.Brain edema expansion was defined as increase on edema grade no less than 2.The unfavorable outcome was defined as the 3-month modified Rankin score>2.Logistic regression was used to identify independent risk factors for HT,edema expansion and unfavorable outcome.Receiver operating characteristics(ROC)curve analysis was used to determine predictive value of rPS for HT,edema expansion and unfavorable outcome.Results:A total of 249 patients were included,among whom the median age was 73(IQR 62-80)years and 101(40.6%)were women.Median baseline NIHSS score was 13(IQR 8-17).Ten patients underwent endovascular therapy alone,57 patients had intravenous thrombolysis(IVT)combined with endovascular therapy and 182 patients had IVT alone.At 24 hours after treatment,HT was observed in 114(45.8%)patients and 86(34.5%)patients had edema expansion.There were 134(53.8%)patients had unfavorable outcome at 3 months after stroke.Logistic analysis showed high rPShypo-i was independently associated with HT(OR=1.009,95%CI 1.003-1.015,p=0.004)and edema expansion(OR=1.007,95%CI 1.002-1.013,p=0.007),whereas high rPShypo-c was independently associated with unfavorable outcome(OR=1.118,95%CI 1.005-1.244,p=0.041).ROC analysis revealed that the optimal rPShypo-i thresholds for HT and edema expansion were 54.47%(sensitivity=78.1%,specificity=49.6%)and 72.51%(sensitivity=73.3%,specificity=63.8%),respectively.For identifying unfavorable outcome,the optimal rPShypo-c threshold was 3.45%(sensitivity=33.6%,specificity=86.1%).Conclusion:AIS patients with higher pretreatment BBBP in hypoperfusion region are more likely to develop HT and brain edema expansion,while patients with higher BBBP in the contralateral mirror hypoperfusion region are more likely to have unfavorable outcome after reperfusion therapy.Part Three The Effect of Reperfusion Therapy on Blood-Brain Barrier Permeability in Patients with Acute Ischemic StrokeAims:Our aims were to detect early changes of the blood-brain barrier permeability(BBBP)in acute ischemic stroke patients who received reperfusion therapy,with or without reperfusion,and find out whether BBBP can predict clinical outcomes.Methods:We retrospectively reviewed our prospectively collected database of AIS patients who received reperfusion therapy within 9 hours of stroke onset and performed head CT perfusion before and after reperfusion therapy,from July 2011 to January 2016.Hypoperfusion region was defined as region with Tmax>6s.Hypoperfusion regions were automatic outlined as regions of interest(ROI)by MIStar in each slice separately,while non-hypoperfusion regions in the ischemic hemisphere were outlined by hand-drawn ROIs,and their mirror ROls in the contralateral nonischemic hemisphere were created across the midline.Then we superimposed these ROIs on the permeability-surface area product(PS)maps and cerebral blood flow(CBF)maps by each slice,respectively,to calculate mean PS and CBF values within each ROIs.We used relative PS(rPS)to represent BBBP.The rPS in hypoperfusion region(rPShypo-i),non-hypoperfusion region(rPSnonhypo-i)and their contralateral mirror regions(rPShypo-c and rPSnonhypo-c)were calculated based on the following formula:rPS=(PS/CBF)×100%.Reperfusion rate(RR)=(baseline hypoperfusion volume-24h hypoperfusion volume)/baseline hypoperfusion volume.We defined reperfusion as ? 70%RR.The unfavorable outcome was defined as the 3-month modified Rankin score>2.The changes of rPS were analyzed using analysis of variance(ANOVA)with repeated measures.Logistic regression was used to identify independent predictors of unfavorable outcome.Results:Fifty-six patients were included in the analysis,within whom 46 patients underwent intravenous thrombolysis(IVT)alone and 10 patients had IVT combined with endovascular therapy.Median age was 76(IQR 62-81)years and 28(50%)were female.Median baseline NIHSS score was 12(IQR 8-15)and median onset-to-needle time was 228(IQR 142-280)min.Thirty-four(61%)patients had 24-hour reperfusion,and 30(54%)patients had unfavorable outcome.From baseline to 24 hours after treatment,rPShypo-i(1.79±0.05 vs 0.67±0.09,p<0.001),rPSnonhypo-i(0.63±0.04 vs 0.48±0.05,p=0.011)and rPShypo-c(0.33±0.05 vs 0.6±0.06,p=0.003)all decreased significantly.After treatment,rPShypo-i(reperfusion group:1.76±0.06 vs 0.28±0.09,p<0.001;non-reperfusion group:1.84±0.08 vs 1.28±0.11,p<0.001)and rPShypo-c(reperfusion group:0.27±0.06 vs 0.121±0.08,p=0.044;non-reperfusion group:0.42±0.08 vs 0.22±0.10,p=0.028)decreased significantly from baseline to 24 hours in both reperfusion and non-reperfusion groups.However,the rPSnonhypo-i decreased from baseline to 24 hours only in reperfusion group(reperfusion group:0.60±0.05 vs 0.36±0.06,p<0.001;non-reperfusion group:0.68±0.06 vs 0.68±0.07,p=0.944),while rPSnonhypo-c did not change from baseline to 24 hours in either group(reperfusion group:0.35±0.04 vs 0.34±0.06,p=0.855;non-reperfusion group:0.43±0.05 vs 0.45±0.07,p?0.788).The rPShypo-i(0.28±0.09 vs 1.28±0.11,p<0.001)and rPSnonhypo-i(0.36±0.06 vs 0.68±0.07,p=0.001)at 24 hours were lower in reperfusion group than in non-reperfusion group.The rPShypo-i at follow up was an independent predictor for unfavorable outcome(OR=1.131,95%CI 1.018-1.256,p=0.022).Conclusion:Early disruption of BBB in AIS is reversible,particularly when greater reperfusion is achieved.Elevated BBBP at 24 hours after treatment,not the pretreatment BBBP,predicts unfavorable outcome. |
PDF Full Text Request