| Objective(s): Early diagnosis and treatment of Alzheimer’s disease(AD)is difficult,and the diagnosis is usually confirmed by cerebrospinal fluid and cranial imaging.The improvement of plasma marker detection technology has made it a convenient and reliable tool to assist in the diagnosis of AD.Nerve filament light chains(NFL)and phosphorylated Tau181 protein(P-tau181)are considered to be the most promising biomarkers for AD.In addition,the presence of Aβ deposits in the retina was found to be a new potential biomarker,confirming that AD brain pathological changes can be manifested through the retina.The purpose of this paper is to investigate the differences in plasma NFL levels,P-tau181 levels,retinal thickness,and vascular density between Mild cognitive impairment(MCI),AD,and normal controls(HC),and whether APOEε4(+)affects NFL,P-tau181 levels,to explore whether the combination of the two fundus examinations can more accurately assist in the diagnosis of AD,and which examinations combine to assist in the diagnosis of AD more accurately.Methods: This study was a cross-sectional study.Subjects of Yunnan origin and Han nationality were collected from the psychiatric outpatient and inpatient departments of the First Affiliated Hospital of Kunming Medical University.Cognitive function was assessed: by the Mini-mental State Examination(MMSE),and Montreal Cognitive Assessment(MOCA).Plasma NFL and P-tau181 levels:Simoa method;Fundus examination: Optical coherence tomography(OCT)and optical coherence tomography angiography(OCTA).APOE gene testing: Sanger method was used.Subjects were divided into three groups,the mild cognitive impairment group,the Alzheimer group,the and normal control group.The levels of NFL and P-tau181 in the three groups were compared,and the relationship between the two and cognitive function was analyzed.To compare whether there are differences in retinal thickness and vascular density among the three groups.To explore which tests can assist in the diagnosis of AD.all statistics were performed using SPSS 25.0.chi-square test,one-way ANOVA,K independent samples t-test,post hoc multiple comparisons,multivariate logistic regression to assess the relationship between NFL and P-tau181 levels and cognitive performance,and ROC curves to evaluate the accuracy of various tests for the diagnosis of AD.Results: A total of 143 subjects were collected,44 with MCI,55 with AD and44 with HC.The results of plasma assay analysis: p NFL levels were higher in the MCI group(17.76(11.04/25.51))than in the HC group(13.09(8.28/17.58)),and there was no significant difference between the MCI and AD(21.66(15.33/31.24))groups.p-tau181 levels were higher in the MCI group(1.87(1.16/3.58))than in the HC group(1.93(1.58/2.37))and higher in AD group(4.29(2.53/5.40))than MCI group.Only subjects in the MCI group with APOEε4(+)had higher P-tau181 levels than those in the APOEε4(-).Adjusting for sex,age,years of education,disease duration,and APOEε4(+),p NFL levels were negatively correlated with MMSE(F =10.279,B=-0.114,R2=0.619,p=0.002),MOCA scores(F =15.611,B=-0.115,R2=0.382,p=0.000),P-tau181 levels were negatively correlated with MMSE(F =9.819,B=-0.517,R2=0.302,p=0.005),and MOCA score(F =15.079,B=-0.732,R2=0.399,p=0.000).Retinal analysis results: RNFL temporal thickness was thicker in the HC group(75.45 ± 13.64)compared to the MCI group(61.06 ± 9.75),and the MCI group(61.06 ± 9.75)had reduced thickness(70.04 ± 9.39)compared to the AD group,with no significant differences between the HC and AD groups.The results of multiple tests combined to assist in the diagnosis of AD: p NFL(AUC=0.712)and P-tau181(AUC=0.763)levels both assisted in the diagnosis of AD,and both were more accurate when combined with fundus examination(AUC=0.784)and more accurate when combined with cognitive scale and fundus examination(AUC=0.983).Conclusion(s): Elevated P-tau181 levels can distinguish MCI from AD.pNFL levels and changes in P-tau181 levels are independent risk factors for cognitive decline.p NFL and P-tau181 levels combined with RNFL temporal thickness are more accurate than the combination of the two to assist in the diagnosis of AD,and the combination of the two with cognitive scales and RNFL temporal thickness has the highest accuracy. |