Analysis Of Correlation Of Red Blood Cell Distribution Width Standrad Deviation With The Severity And Location Of Cerebral Microbleeds

ObjectiveExplore the correlation of red blood cell distribution width standrad deviation(RDW-SD)with the severity and location of cerebral microbleeds(CMBs)through retrospective analysis.Methods and materialsThe data were retrospectively collected from 103 patients admitted to the Department of neurology of our hospital between January 2020 and November 2022,who were confirmed the presence of CMBs in S WI sequence and met the inclusion and exclusion criteria.The clinical indicators include age,gengder,with or without hypertension,with or without diabetes,with or without smoking history,with or without drinking history.The blood indicators include red blood cell count,red blood cell distribution width standrad deviation,hemoglobin,white blood cell count,platelet count,total cholesterol,creatinine,high density lipoprotein,uric acid,low density lipoprotein,triglycerides and homocysteine.Two clinically experienced neurologists using double-blind method based on the brain observer microbleed scale(abbreviated as BOMBS)and microbleed anatomical rating scale(abbreviated as MARS)to determine the severity and location of CMBs by SWI sequence results and the patients were divided into different groups.Then the risk factors were analyzed and the predictive value of relevant indicators was drawn by plotting the receiver operating curve(ROC curve).It was considered that P<0.05 was significant.Results1.The results of risk factors analysis for lobar CMBsThere were 20 patients in the lobar CMBs group and 32 patients in the non-lobar CMBs group.The results of univariate analysis showed that,compared with the nonlobar CMBs group,the prevalence of diabetes(P=0.018)and the RDW-SD levels(P<0.001)was lower in lobar CMBs group.The binary logistic regression model was applied to analyse the independent variables of P<0.1 in the univariate analysis,the results showed that there was a significant correlation between the diabetes and lobar CMBs.Decreased prevalence of diabetes(OR=0.107,95%CI:0.014～0.818,P=0.031)might be an independent risk factor for lobar CMBs after adjusting for confounding factors.2.The results of risk factors analysis for deep CMBsThere were 20 patients in the deep CMBs group and 33 patients in the non-deep CMBs group.The results of univariate analysis showed that,compared with the nondeep CMBs group,the RDW-SD(P=0.044),the hemoglobin and uric acid levels(P=0.015 and P=0.009)were higher in deep CMBs group.The binary logistic regression model was applied to analyse the independent variables of P<0.1 in the univariate analysis,the results showed that nothing were found to be independent risk factors for deep CMBs after adjusting for confounding factors.3.The results of risk factors analysis for the severity of CMBsAmong 103 CMBs patients,28 cases were mild CMBs and 75 cases were moderate-severe CMBs.The results of univariate analysis showed that,compared with mild CMBs group,the prevalence of hypertension was higher(P=0.018),the RDW-SD and uric acid levels(P=0.038 and P=0.043)were higher,while the low density lipoprotein levels(P=0.012)were lower in moderate-severe CMBs group.The binary logistic regression model was applied to analyse the independent variables of P<0.1 in the univariate analysis,the results showed that there was a significant correlation between the RDW-SD,low density lipoprotein and moderate-severe CMBs.Increased RDW-SD levels(OR=1.321,95%CI:1.098～1.590,P=0.003)and decreased low density lipoprotein levels(OR=0.395,95%CI:0.215～0.740,P=0.003)might be independent risk factors for moderate-severe CMBs after adjusting for confounding factors.4.Correlation between the severity of CMBs and the factors affecting the severity of CMBsAccording to the results of univariate analysis,the independent variables of P<0.05 were included in Spearman correlation analysis,which showed that the severity of CMBs was weakly positively related to RDW-SD(r=0.341,P=0.000)and the prevalence of hypertension(r=0.233,P=0.018),respectively,while weakly negatively related to low density lipoprotein levels(r=-0.221,P=0.025).5.Plot the receiver operating curve of RDW-SD between moderate-severe CMBsThe area under ROC curve of RDW-SD was 0.721(95%CI:0.610～0.831,P=0.001).The best cut-off value of 43.85 when the Youden index was at its maximum,with a sensitivity of 53.60%and specificity of 84.00%for the prediction of moderatesevere CMBs.ConclusionsThere was a significant correlation between the RDW-SD and moderate-severe CMBs.Increased RDW-SD levels might be an independent risk factor for moderatesevere CMBs.