Glibenclamide Advantage In Treating Edema After Intracerebral Hemorrhage (GATE-ICH)

【Background】As the most critical factor of secondary brain injury,perihematomal edema(PHE)after primary intracerebral hemorrhage(ICH)could further aggravate the neurological deterioration.At present,traditional dehydration therapy has been the most common option in treating edema in the clinical practice.However,the effectiveness of traditional dehydration therapy is modest.The sulfonylurea receptor 1-transient receptor potential melastatin 4 after the injury of central nervous system could be inhibited by glibenclamide,which may contribute to reduce both cytotoxic and vasogenic edema.The effect of glibenclamide for treatment of brain edema has been verified in previous clinical studies about ischemic stroke.It was also suggested that glibenclamide may inhibit the formation of hematoma and PHE,as well as the functional outcomes after ICH.Therefore,glibenclamide could be regard as a new hope for treatment of ICH.【Objectives】To explore the effect of glibenclamide on the formation of PHE and functional outcomes after ICH.The contents were divided into three parts.1.To retrospectively analyze the effect of SFUs on the volume of PHE in patients with type 2 diabetic and the clinical outcome after ICH.2.A single center prospective study was conducted to explore the effect of glibenclamide in preventing PHE and clinical outcomes after ICH,so as to provide the evidence for a prospective multicenter clinical trial.3.A multicenter prospective study was conducted to explore the effect of glibenclamide in the treatment of PHE and 90-day functional prognosis after ICH.【Methods】1.Research 1: A single center retrospective clinical study From January 2009 to October 2018,we conducted a retrospective case-control study to analyze the patients with ICH locating at basal ganglia accompanying type 2 diabetes in Xijing Hospital.According to the established matching factors,the patients who used SFUs before ICH were matched twice as many as those who did not use SFUs.The baseline clinical data,hematoma volume and edema volume on head CT,as well as clinical outcomes of the two groups at admission were collected.The brain edema was assessed by PHE and r PHE on head CT,while the clinical outcomes was assessed by modified Rankin Scale(m RS)and Barthel index at discharge.The differences of edema volume and clinical outcomes were compared between the two groups.Linear regression analysis and binary logistic regression analysis were used to explore whether the use of SFUs before onset in patients with ICH could affect the formation of PHE and clinical outcome at discharge.2.Research 2: A single center prospective clinical study We prospectively enrolled 24 patients with acute ICH locating basal ganglia admitted to Department of Neurology in Xijing Hospital.The patients randomized into glibenclamide group were treated with glibenclamide 1.25 mg,3/day and usual background care,while the patients randomized into control group were only treated with usual background care.The primary outcome of this study was the volume of PHE at day7.The difference of hematoma volume,PHE volume and r PHE at day 7 after enrollment,as well as the functional outcomes at day 7 and day 90 after enrollment between two groups were also analyzed.3.Research 3:A multicenter prospective clinical study A prospective,multicenter,randomized,controlled clinical study,Glibenclamide Advantage in Treating Edema after Intracerebral Hemorrhage(GATE-ICH)trial,was conducted in Xijing Hospital and 25 other hospitals in Shaanxi province.220 patients with acute ICH were randomized into glibenclamide group and control group.According to the pilot study,patients assigned to the glibenclamide group were given glibenclamide 1.25 mg,3/day and usual background care,while patients assigned to the control group were given the same background care only.The primary outcome of the GATE-ICH trial was the proportion of patients with poor outcome(the score of m RS more than 3).The secondary outcomes included the changes of hematoma volume,brain edema on CT scans at day 3 and day 7,as well as the clinical scores at day 3,day 7 and day 90.The brain edema was evaluated by 6 measurements including absolute PHE,r PHE,edema extension distance(EED),peak PHE,the increase and growth rate of PHE.The clinical scores were evaluated by 3-day and 7-day National Institutes of Health Stroke Scale(NIHSS)and Glasgow Coma score(GCS),as well as m RS and Barthel index at day 3,day 7 and day90.【Results】1.Research 1:We enrolled a total of 27 patients with ICH locating basal ganglia in the first study,consisting of 9 patients in SFU group and 18 patients in non-SFU group.Compared to patients assigned to the non-SFU group,the time from onset to the first CT scan in patients assigned to the SFU group was significantly longer(1 [1-1] vs.1 [1-3] days,p=0.011).The results showed that the hematoma volume of the two groups was comparable.The non-SFU group presented with larger PHE volume and r PHE than the SFU group(PHE: 15.4 [7.4-50.2] m L vs.8.0 [3.1-22.1] m L,p= 0.100;r PHE: 1.5 [1.2-1.9]vs.0.8 [0.7-1.3],p=0.006).No significant difference was found in the score of m RS and Barthel index,as well as the proportion of patients with m RS ≥ 3 and mortality between the two groups.After adjusting for the confounding factors,SFUs could significantly reduce the PHE volume(coefficient:-13.607,95%CI:-26.185--1.029,p=0.035)and r PHE(coefficient:-0.566,95%CI:-0.971--0.161,p=0.009),but had no significant effect on the clinical outcomes.2.Research 2:From January 2018 to September 2018,we enrolled 24 patients and 2 patients were withdrawn.The final analysis included 22 patients,consisting of 10 patients in glibenclamide group and 12 patients in the control group.Most of the baseline data wes comparable between the two groups,such as age,baseline NIHSS score,ICH Score,hematoma volume,PHE and r PHE,excepting for the time from onset to randomization(glibenclamide group: 35.0 [23.3-48.0] h vs.control group: 19.0 [12.2-24.0] h,P=0.030).At day 7,patients in glibenclamide group had significantly lower PHE volume than the control group(21.74 ± 9.70 m L vs.31.22 ± 10.16 ml,p=0.038).No significant difference was found in the clinical scores at baseline or at discharge.However,patients in glibenclamide group had significantly improvement of Barthel index than control group(20.0 [10.0-21.3] vs.5.0 [0.0-15.0],p=0.017).The patients in glibenclamide group presenting with higher scores of 90-day Barthel index than control group(90.0 vs.80.0,p=0.016).The scores of m RS from baseline to day 90 decreased significantly in glibenclamide group than that of control group(3.0 [2.0-3.0] vs.2.0 [1.0-2.0],p=0.007).After adjusting for the confounding factors,glibenclamide usage could significantly reduce the PHE volume(coefficient:-12.908,95%CI:-22.271--3.665,p=0.014),but had no significant effect on the r PHE and clinical outcomes.The safety events between the two groups were similar.3.Research 3:A total of 220 patients were enrolled in the GATE-ICH trial.From December 2018 to September 2020,we screened a total of 1537 patients in 26 hospitals,of whom 220 patients underwent randomization.We excluded 20 subjects who withdrew their consent or not eligible for the GATE-ICH trial.The final 200 patients included 99 patients in the glibenclamide group and 101 patients in the control group.The baseline data was similar,such as age,time from onset to randomization,baseline hematoma volume and PHE volume,as well as NIHSS score and ICH Score between the two groups.The incidence of90-day poor outcome(m RS ≥ 3)in glibenclamide group was 20.2%,which was 9.5%lower than that in control group(29.7%)(p=0.121).The adjusting glibenclamide usage has the trend of correlation to the low shift of 90-day m RS(OR: 0.590,95%CI:0.345-1.009,p=0.054).No significant difference was found in other clinical secondary outcomes.The volume of hematoma and PHE was comparable between glibenclamide group and control group at baseline,3 days and 7 days.However,the expansion of PHE from day 3 to day 7 in glibenclamide group was significantly lower than that in control group(2.9 [-0.1-6.8] m L vs.5.5 [0.7-10.9] m L,p = 0.025).After adjusted the confound factors,glibenclamide usage could reduce the PHE volume(coefficient:-4.482,95%CI:-7.619--1.344,p=0.005),EED(coefficient:-0.073,95%CI:-0.137--0.010,p=0.023)at day 7,and peak PHE(coefficient:-3.012,95%CI :-5.661--0.414,p=0.023).Furthermore,the using of glibenclamide could significantly reduce the increase of PHE volume(from day 3 to day 7: coefficient:-3.752,95%CI:-6.337--1.166,p=0.005;from baseline to day 7:-3.686,95%CI:-6.809--0.563,p=0.021).As for the safety events,the proportion of asymptomatic hypoglycemia in glibenclamide group was significantly higher than that in control group(0.0% vs.15.2%,p<0.001).There were 3(3.0%)subjects in glibenclamide group and none(0.0%)in control group presented with symptomatic hypoglycemia(p=0.119).No significant difference was observed in other safety events between glibenclamide group and control group.【Conclusions】1.In patients accompanying type 2 diabetes,the volume of PHE was smaller in patients using SFUs than those not using SFUS before ICH.However,the clinical outcomes at discharge were similar in the groups.2.The using of low-dose glibenclamide in the acute stage of ICH could significantly alleviate the formation of PHE and might indicate better clinical outcomes.3.The low-dose glibenclamide usage after ICH could significantly inhibit the formation of PHE and indicate low shift of 90-day m RS.Although the using of glibenclamide could reduce the incidence of poor outcome by 9.5%,it has no statistical significance.Compared to the control group,the probability of hypoglycemia in glibenclamide group was significantly higher.