Study On The Relationship Between The Spatial Structure And Distribution Characteristics Of Hematoma And The Expansion Of Early Hematoma

Objects:Hematoma growth after intracerebral hemorrhage(ICH)is considered an interventional therapeutic target that correlated with the outcome after ICH.Therefore it is of great significance to improve the prognosis after ICH by early identifying the patients who have higher risk in developing hematoma growth,and consequently implementing active treatment.At present,the imaging index applied to predict hematoma growth are all based on the evaluation of hematoma morphology or concentration on a 2-dementional image,which limits the accuracy of management.However with the help of professional medical image processing software,it is feasible to assess several characteristics of the brain tissue and hematoma based on a 3-dementional level.In the current study,we quantitatively measured the ICH hematoma morphology,distribution and the characteristics of the surrounding structures via a professional medical image processing software,and explored the association between the recorded index and hematoma growth.Methods:The demographic,clinical and imaging information were collected from 127 patients with spontaneous ICH from the cerebral small vessel disease cohort study of Peking Union Medical College Hospital.For all the enrolled patients,the first CT scan was performed within 6 hours since onset,and a re-examination of CT scan was done within the following 48 hours.In all patients,the indices were measured by medical image processing software including the volume of hematoma,the surface area of hematoma,surface area/volume ratio(SVR),distribution of the hematoma volume,volume of the cerebrospinal fluid,and volume of the brain tissue.Hematoma growth was classified as an increase in hematoma volume of ? 12.5ml or 33%compared with baseline.The hematoma volume,surface area,SVR,volume of the brain tissue,volume of the cerebrospinal fluid,volume of cerebrospinal fluid/brain tissue,and the differences in various hematoma distribution were all calculated by univariable and multivariable analysis.And the association between the abovementioned index and hematoma growth was also evaluated.Results:Among all the enrolled 127 cases with spontaneous ICH,totally 27 patients(21.26%)experienced hematoma growth.The SVR on baseline CT scan was higher in patients with hematoma growth compared to those without hematoma growth[2.62(2.49,3.94)cm2/ml vs.2.43(2.25,2.74)cm2/ml,p<0.001].Based on multivariable Logistic regression analysis,higher SVR was independently associated with hematoma growth(OR:6.538,95%CI:2.147-19.908,p=0.001).In the prediction of hematoma growth by SVR,the area under the receiver operating characteristic curve was 0.732,and the sensitivity and specificity were 51.0%and 88.9%respectively.In patients with hematoma growth,the re-examined CT scan revealed lower SVR compared to the baseline CT scan[2.59(2.12,3.21)cm2/ml vs.2.62(2.49,3.94)cm2/ml,p=0.002].Besides the re-examined CT scan demonstrated that the volume of the hematoma in the frontal lobe,the parietal lobe,the temporal lobe,the basal ganglia and the brain stem increased compared to baseline CT scan in patients with hematoma growth(p<0.05).No significant association was found between distribution of the hematoma volume and hematoma growth.No significant association was found between hematoma growth and theb volume of cerebrospinal fluid,the volume of brain parenchyma and the ratio of cerebrospinal fluid/brain parenchyma volume.Conclusion:According to the quantitative analysis on the SVR of hematoma,it was found that irregular morphology was an independent risk factor of hematoma growth.The baseline SVR of hematoma may be an imaging index for quantitatively predicting the occurrence of hematoma growth.Furthermore hematoma growth was correlated with decreased SVR of the hematoma growth,indicating that surface tension may be an underlying mechanism of hematoma growth.