Study On The Association Between Rate Of Ultraearly Hematoma Growth,hematoma Shape And Early Hematoma Expansion In Patients With Intracerebral Hemorrhage

PART ONE STUDY ON RATE OF ULTRAEARLYHEMATOMA GROWTH AND EARLY HEMATOMAEXPANSION IN SPONTANEOUS INTRACEREBRALHEMORRHAGE IN DIFFERENT TIME COURSEBackground Rate of ultraearly hematoma growth(RUHG)is a useful predictor of hematoma expansion(HE).The purpose of our study was to identify the optimal value of RUHG for predicting HE and to investigate whether the effect of RUHG on HE was time-dependent.Material and methods Patients with ICH who underwent CT scan within 6 h of symptoms onset and follow-up CT scan within 36 h after the initial CT scan were included in our study.ABC/2 method was used to measure hematoma volume on the primary CT imaging.The RUHG was defined as hematoma volume(m L)/onset-to-imaging time(hours).Multivariate logistic regression analyses were used to identify the association between clinical variables and HE.Areas underreceiver-operating characteristic curves were used to determine the optimal RUHG and baseline hematoma volume for HE in patients presented within0-3 h and 3-6 h.Results In 257 patients with ICH of this cohort study,there were 85(33.1%)patients with HE.Median RUHG with HE was significantly higher than for non-HE(11.7 [6.4-29.1] vs.4.7 [2.3-9.3] m L/h;P = 0.001).Multivariate logistic regression model revealed that RUHG was independently associated with HE.The optimal value of RUHG for predicting HE was > 6.1 m L/h in patients presented within 6 h after onset of symptoms.Then optimal values of RUHG in patients with onset-to-imaging time ? 3 h and onset-to-imaging time 3-6 h were 9.1m L/h and 5.1 m L/h,respectively.Conclusions RUHG independently predicts HE in patients with ICH,but predicting HE better in time course of 0-3 h from onset to imaging.PART TWO COMPARATION OF RATE OFULTRAEARLY HEMATOMA GROWTH AND FORPREDICTING HEMATOMA EXPANSIONObjective To investigate the relationship between rate of ultraearly hematoma growth(RUHG)and hematoma expansion(HE)in patients with intracerebral hemorrhage(ICH)among difference subgroups.Patients and methods A cohort study was analyzed retrospectively among 251 eligible data of Chinese patients with ICH.Regions of interest(ROI)on target images were measured by computer-assisted planimetry.Clinical characteristics and image features on baseline were investigated for HE and non-HE.RUHG were processed by Multi-logistic regression model for HE among different Subgroups.predictive value of RUHG for HE was calculated by ROC curves and plotted for comparison among different subgroups.Results SBP,GCS score,time from onset-to-baseline CT,time from initial CT to followed CT,baseline hematoma volume,RUHG and blend sign were associated with HE(p < 0.05).Multi-logistic regression analysis showed that RUHG was an independent risk factor with HE among subgroups of age ? 60 year and <60 year,male and female,SBP ? 180 mm Hg,deep hematoma location.AUC of RUHG for predicting HE was better in subgroups of age ? 60 year(0.812),female(0.788),SBP ? 180 mm Hg(0.788),deep hematoma location(0.753).Comprehensive comparison with sensitivity,specificity,PPV,NPV,and AUC showed the same result of AUC comparation.Conclusions Relationship of RUHG with HE was complicated.Analysis with RUHG for predicting HE among subgroups of age,sex,SBP and hematoma location might be necessary for clinical strategy.PART THREE INTER-GROUP STUDY ONMORPHOLOGYAND HETEROGENEITY,EARLYHEMATOMA EXPANSIONObjective To evaluate correlation of hematoma shape with assessment of hematoma expansion and heterogeneity based on ABC/2 method and Computer-assisted planimetry.Patients and methods there were 240 eligible patients with ICH were investigated for clinical characters and imaging feature on baseline.the shape of baseline hematoma was classified into regular,irregular-lightly,irregular-heavily and separated shape.The percent of different imaging markers of ICH were compared among different shapes.ABC/2 method and planimetry were compared for the consistence of assessing HE,volume,rate of ultraearly hematoma growth(RUHG)respectively.ABC/2 method for predicting HE was evaluated by ROC curves.Hazard ratio of hematoma shapes and RUHG for predicting HE was processed by multi-logistic regression model.Results Age,GCS scores and ICH density,midline shift were significantly difference among different subgroups of hematoma shapes(p< 0.05).The percentage of island sign and swirl sign was highly associated with irregular-heavily and separated shape(p < 0.05).Blend sign emerged significantly with irregular-lightly shape(p < 0.05).Consistencecomparation for assessing HE showed the best for regular shape,the lowest for irregular-heavily shape.AUC of ABC/2 method for predicting HE was similar in irregular-lightly shape and in irregular-heavily shape(0.884 vs.0.889),but just higher than in separated shape(0.826).Separated shape was associated with HE according to ABC/2 method and planimetry assessing.Irregular-lightly and irregular-heavily shape were just associated with HE based on planimetry assessing.RUHG was associated with HE in subgroups of irregular-lightly shape,but not in subgroups of separated shape based on ABC/2 method and planimetry assessing.RUHG was associated with HE in subgroups of irregular-heavily shape only based on planimetry assessing.Conclusions Hematoma shapes is an useful factor for assessing HE and heterogeneity.