Correlation Between Deep Medullary Vein Score And Post-stroke Cognitive Impairment In Patients With Acute Ischaemic Stroke

Objectives:To investigate the correlation between the deep medullary vein(DMV)score and post-stroke cognitive impairment(PSCI)in patients with minor acute ischemic stroke(AIS)hoping to provide new insights for the early diagnosis of PSCI.Research objects and methods:We enrolled 219 patients with minor AIS who were hospitalized at the Department of Neurology during the period of April 2019 ～ December 2021 at baseline,collected general clinical data,laboratory data,imaging data and neuropsychological data,applied sensitive weighted imaging(SWI)and DMV score to assess DMV continuity and visibility.Patients were followed up with neuropsychological assessment 6 months after stroke,and 113 patients with complete clinical data and completed follow-up were analyzed.The patients were divided into a PSCI group(47 patients)and a post-stroke no cognitive impairment(PSNCI)group(66 patients)based on the Montreal cognitive assessment(Mo CA)scores during the follow-up period.Firstly,linear regression analysis was applied to investigate the factors influencing DMV at baseline;Secondly,the variability of clinical data,laboratory tests and routine MRI markers and DMV scores at baseline were compared between groups.Then,the receiver operating characteristic curve(ROC)was applied to analyze the predictive value of DMV alone and the model with the rest of the influences on PSCI;Finally,the relationship between DMV scores and cognitive level of each cognitive domain was analyzed by linear regression model.Results:1.A total of 219 patients were enrolled in this study at baseline,113 of whom had complete clinical information and completed follow-up.The mean age of the 113 patients was 61.72 ± 7.27 years,the mean DMV score was 7.9 ± 4.6,and 47 patients developed PSCI,with a prevalence of 41.5%.2.DMV score was positively correlated with age(P < 0.05),negatively correlated with hemoglobin(P < 0.05),and positively correlated with c-reactive protein(CRP)and previous stroke history(P < 0.05 after correcting for age and hemoglobin).DMV score was not associated with smoking history,alcohol consumption,hypertension,diabetes mellitus,uric acid,glycosylated hemoglobin(Hb A1c),lipids,homocysteine(Hcy),folic acid,vitamin B12,and APOE ε4genotype(P > 0.05 after correcting for age and hemoglobin).3.There were differences between the PSCI and PSNCI groups in terms of years of education,age,uric acid,cerebral small vessel disease(CSVD)score,cerebral microbleeds(CMBs),Fazekas score,and DMV score(P < 0.05).There were no statistical differences in previous stroke,hypertension,diabetes mellitus,smoking history,alcohol history,strategic infarct locations,diameter of maximum transverse section DMTS,lacunar space,APOE ε4 genotype,systolic blood pressure,diastolic blood pressure,Hb A1 c,all lipids,Hcy,folic acid,vitamin B12,CRP between the two groups(P > 0.05).4.Multifactorial logistic regression analysis showed that high DMV score and high CSVD score were independent risk factors for the occurrence of PSCI and high education was an independent protective factor for occurrence of PSCI(P < 0.05).5.The ROC curve analysis revealed that a DMV score >9 was the best cut-off value for predicting the occurrence of PSCI.In addition,combining the DMV score with years of education and CSVD score improved the predictive value of the model for PSCI.6.DMV scores were negatively correlated with Mo CA and auditory verbal learning test(AVLT)scores(P < 0.05 after correcting for age,years of education,previous stroke history,hypertension,diabetes mellitus,critical site cerebral infarction,and CSVD scores).DMV scores were not correlated with the mini-mental state examination(MMSE),BOSTON naming test,the verbal fluency test(VFT),Trail Making Test(TMT)scores,or visuospatial scores(P > 0.05 after correcting for age,years of education,previous stroke history,hypertension,diabetes mellitus,critical site cerebral infarction,and CSVD scores).Conclusion:1.Advanced age,previous stroke,high CRP,and low hemoglobin were associated with high DMV scores,and may be influential factors in the low continuity and visibility of DMV.2.Years of education,age,uric acid,CMBs,Fazekas score,DMV score,and CSVD score were associated with the occurrence of PSCI,and high CSVD score and DMV score were independent risk factors for the occurrence of PSCI,and high education was an independent protective factor for occurrence of PSCI.3.DMV score is expected to be a diagnostic imaging biomarker for PSCI,and DMV score >9 at baseline is the best predictor of PSCI in patients with mild AIS.4.In patients with mild AIS,high DMV scores were associated with global cognitive and memory decline in the distant future.The DMV score can be applied to reflect changes in long-term cognitive function in patients with minor AIS.