Association Between Cognitive Function With Ankle-brachial Index And Imaging Features In Patients Of MCI-AD And AD

Objective:In the first part of this paper,the differences in ABI,MRI indicators for CSVD and brain atrophy across the NC,MCI-AD and AD groups are explored,and the optimal variables representing MCI-AD and/or AD are identified using forward logistic regression;the second part firstly analyzes the relationship between ABI and the MRI indicators for CSVD and brain atrophy.Secondly,the correlation between ABI and cognitive test scores was analyzed through neuropsychological testing,and further discusses its relationship with specific cognitive domains.Finally,ROC analysis confirmed the diagnostic value of ABI in MCI-AD and/or AD.Method:There were 27 cases of MCI-AD patients and 23 cases of AD patients collected from the outpatient and department of neurology of the Affiliated Hospital of Soochow University from December 2016 to February 2018,which were in line with relevant diagnostic criteria.Meanwhile,27 cases of healthy volunteers were recruited with normal cognitive function,and exclusion of depression,systemic chronic diseases that cannot be controlled by standard treatment and used drugs that affected cognitive function.All the subjects were tested by neuropsychological scales after being informed consent,these scales included the Mini-Mental State Examination(MMSE),Montreal Cognitive Assessment(Mo CA),Activity of Daily Living Scale(ADL),Clinical Dementia Rating Scale(CDR),Hachinski Ischemic Score(HIS),Hamilton Depression Scale(HAMD).Blood pressure cuff and portable doppler flow detector were used to measure the ABI,and the ABI values were calculated using the method which were recommended by American Heart Association(AHA).After ruling out the relevant taboo,all subjects were underwent 3.0T MRI scan.The MRI indicators of CSVD were independently evaluated by two senior imaging physicians using a double-blind method.The Kappa test was used to analyze the consistency of the two evaluation results.The MRI indicators of brain atrophy were performed on the phase of T1 Flair,and the hippocampal furrow ratio was calculated according to the method of Saka et al.The caudate nucleus index,the frontal horn index,the frontal interhemispheric fissure ratio and the sylvian fissure ratio were calculated by referring to the methods of Hamano et al.and Gomori et al.Using SPSS 19.0 statistical software data processing,the ABI,MRI indicators and cognitive test scores among the three groups were compared and analyzed.Results:1.Comparison of clinical characteristics and cognitive scores among the three groups(1)Age:NC group<AD group,P<0.05;education:NC?MCI-AD group>AD group,P<0.05.(2)Periventricular hyperintensity(PVH)score:NC group<MCI-AD group,P< 0.05;NC group<AD group,P<0.01;deep subcortical white matter hyperintensity(DSW MH)score:NC group<AD group,P<0.01;MCI-AD group<AD group,P<0.05;centrum semiovale perivascular spaces(CSO-VRS)score:NC group<AD group,P<0.05;basal ganglia perivascular spaces(BG-VRS)score:NC group<AD group,P<0.05.(3)ABI:NC group<MCI-AD group,P<0.01;NC group<AD group,P<0.01;hippocampal sulcus ratio:NC group<MCI-AD group<AD group,P<0.05;caudate nucleus index:NC group<MCI-AD group<AD group,P<0.05;frontal horn index:NC group<AD group,P<0.01;MCI-AD group<AD group,P<0.05;frontal interhemispheric fissure ratio:NC group<AD group,P<0.01;MCI-AD group<AD group,P<0.01;Sylvian fissure ratio :NC group<AD group,P<0.01;MCI-AD group<AD group,P<0.01.(4)MMSE?Mo CA scores:NC group>MCI-AD group>AD group,P<0.01;ADAS-Cog score:NC group<MCI-AD group<AD group,P<0.01;ADL score:NC group<AD group,P<0.01;MCI-AD group<AD group,P<0.01.2.Associations between ABI with MCI-AD and/or AD(1)Logistic regression analysis shows that ABI decline is related to MCI-AD(OR=2.868,95%CI:1.405-5.854,P=0.004),AD(OR=2.745,95%CI:1.349-5.587,P=0.005),MCI-AD and AD(OR=2.641,95%CI:1.481-4.710,P=0.001).Adjusting for age and education,ABI decline is still related to MCI-AD(OR=3.014,95%CI:1.443-6.296,P= 0.003)?AD(OR=3.015,95%CI:1.303-6.979,P=0.005)?MCI-AD and AD(OR=3.055,95%CI:1.568-5.954,P=0.001).(2)The logistic regression analysis of the forward method shows that:for MCI-AD,ABI(OR=3.957,95%CI:1.687-9.281,P=0.002)and hippocampal sulcus ratio(OR=0.354,95%CI:0.165-0.759,P=0.008)were retained in the model;for AD,ABI(OR=6.45 7,95%CI:1.354-30.784,P=0.019)and sylvian fissure ratio(OR=0.225,95%CI:0.103-0.491,P=0.000)were retained in the model;for MCI-AD and AD,ABI(OR=3.369,95%C I:1.583-7.172,P=0.002)and caudate nucleus index(OR=0.225,95%CI:0.103-0.491,P=0.000)were retained in the model.3.Associations between ABI and cognitive test scores and specific cognitive domains(1)Multiple linear regression analysis showed that in the fully adjusted model,the total scores of MMSE(b=0.251,95%CI:0.093-0.408,P=0.002),Mo CA(b=0.235,95%CI: 0.084-0.386,P=0.003),and ADAS-Cog(b=-0.198,95%CI=-0.348--0.049,P=0.010)varied with changes in ABI.(2)In the fully adjustment model of specific cognitive domains,a.For the MMSE subitems,the computational power(b=0.315,95%CI:-0.120-0.509,P=0.002)and language ability(b=0.281,95%CI:0.109-0.453,P=0.002)scores varied with changes in ABI;b.For the Mo CA subitems,visual space/execution function(b=0.184,95%CI:0.017-0.350,P=0.031),attention(b=0.300,95%CI:0.116-0.485,P=0.002),language ability(b=0.266,95%CI:0.092-0.441,P=0.003)scores varied with changes in ABI;c.For ADAS-Cog subitems,word recall(b=-0.239,95%CI:-0.399--0.078,P=0.004),recall test(b=-0.193,95%CI:-0.373--0.012,P=0.03 7),verbal expression(b=-0.268,95%CI:-0.471--0.066,P=0.010),and language understanding(b=-0.294,95%CI:-0.460--0.127,P=0.001)scores varied with changes in ABI;d.For ADL subitems,physical life self-care(b=-0.250,95%CI:-0.437--0.064,P=0.009)scores varied with changes in ABI.4.The ROC analysis of ABI in MCI-AD and/or ADThe area under the curve of ABI for MCI-AD,AD,MCI-AD and AD was 0.767,0.740,and 0.760,respectively;the best cut-off point for diagnosis of MCI-AD and/or AD was 1.095;the sensitivity of diagnosis of MCI-AD,AD,MCI-AD and AD were all 81.50%;the specificities were 66.70%,65.20%,and 64.00%,respectively;the positive predictive values were 79.17%,73.68%,and 86.84%,respectively;negative predictive values were 73.33%,70.97%,and 56.41%,respectively;the accuracy were 75.93%,72.00%,and 71.43%,respectively.Conclusion:1?Age,low education is a risk factor for AD;ABI decline can be used as an early predictor of AD;ABI combined with white matter hyperintensity,perivascular space score,and various magnetic resonance of brain atrophy can be used as a tool to assess the severity of AD;cognitive function tests can help identify NC,MCI-AD,and AD.2?The decrease of ABI is related to MCI-AD and AD.The ABI and typical MRI brain atrophy indexes in each period are the best alternative indicators for MCI-AD and AD.3?The decrease in ABI is related to the decline in total cognitive rate.It is related to computational power,language ability,visual space/executive function,visual memory,and physical life self-care function in the specific cognitive domain,especially in digital processing.4?The best cut-off point of ABI for diagnosis of MCI-AD and AD is 1.095,which has high sensitivity,poor specificity and a certain accuracy.