【摘要】 目的 探究基于光学相干断层扫描(OCT)及光学相干断层扫描血管造影(OCTA)技术的视网膜标志物在认知损害诊断中的应用，并分析其与人口学特征及共患病之间的关联性。方法 纳入来自上海记忆研究和上海老年研究的受试者374例，经临床诊断分为认知损害组和认知正常组，采集两组人口学及共患病信息，应用OCT及OCTA技术检测其视网膜厚度及血管指标，探索在认知损害诊断中人口学信息及共患病对视网膜标志物的影响。结果 认知损害组视网膜神经纤维层(RNFL)、神经节细胞复合体(GCC)、黄斑区视网膜全层及内层厚度和视网膜浅层及深层毛细血管丛的血管密度与认知正常组比较差异均有显著性(均P<0.05)。年龄对视网膜厚度及血管密度存在显著影响(r=-0.456～-0.123,P<0.05)，经校正性别后仍呈相关性(均P<0.05)。校正性别后，认知正常组视网膜血管密度、认知损害组视网膜厚度和浅层血管密度与年龄均呈相关性(均P<0.05)。女性与更薄的黄斑区视网膜厚度和更大的深层血管密度呈相关性(均P<0.05)；校正年龄后，认知正常组女性的黄斑区视网膜厚度显著更薄(r=-0.312～-0.196,P<0.05)，而认知损害组女性的深层血管密度显著更高(r=0.224～0.264,P<0.05)。脑卒中病史是视网膜指标的重要影响因素，糖尿病病史对认知正常组的RNFL及GCC厚度产生影响。结论 视网膜厚度及血管参数有潜力作为非侵入性、易获取的认知损害诊断标志物，但实际应用需考虑人口学特征和共患病等影响因素，以提高准确性和可重复性。更多还原

【Abstract】 Aim To investigate the application of retinal biomarkers based on optical coherence tomography(OCT) and optical coherence tomography angiography(OCTA) technology in the diagnosis of cognitive impairment, and to analyze their correlations with demographic characteristics and comorbidities. Methods A total of 374 subjects from the Shanghai Memory Study and the Shanghai Aging Study were included, clinically diagnosed, and classified into cognitively impaired(CI) group or cognitively normal(CN) group. Demographic and comorbidity information was collected, and OCT and OCTA technology were used to detect retinal thickness and vascular biomarkers. The effects of demographic and comorbidity factors on retinal biomarkers in the diagnosis of cognitive impairment were explored. Results There were significant differences in the thickness of retinal nerve fiber layer(RNFL),ganglion cell complex(GCC), full and inner layer thickness of macular retina, and vascular density of superficial and deep capillary plexuses between the CI group and the CN group(P<0.05). Age and sex were important factors influencing retinal biomarkers. Age had a significant and widespread impact on retinal thickness and vascular density(r=-0.456～-0.123,P<0.05), and this correlation remained significant after adjusting for gender(all P<0.05). After adjusting for gender, there was a significant correlation between retinal vascular density in the CN group and retinal thickness and superficial vascular density in the CI group with age(all P<0.05). Female was significantly associated with thinner macular retinal thickness and higher deep vascular density(all P<0.05), and after adjusting for age,female participants in the CN group still had significantly thinner macular retinal thickness(r=-0.312～-0.196, P<0.05), while female participants in the CI group had significantly higher deep vascular density(r=0.224～0.264, P<0.05). A history of stroke was an important factor influencing retinal indicators, while a history of diabetes affected RNFL and GCC thickness in CN group. Conclusion Retinal thickness and vascular parameters are promising to serve as non-invasive and easily obtainable diagnostic biomarkers for cognitive impairment. However, in practice, demographic factors and comorbidities of patients need to be taken into consideration to improve the accuracy and repeatability of these biomarkers.更多还原