Table 3.
Associations of neuronal and microvascular structures and function of the neurovascular coupling unit, as quantified by retinal imaging, with cognitive performance.
| N | Cognitive performance assessment | Confounders | Main results | Reference |
|---|---|---|---|---|
| Retinal nerve fiber layer thickness | ||||
| Prospective studies | ||||
| N=1,251 (3-years follow-up) |
Global cognitive performance was estimated from the following tests: prospective memory, pairs matching, numeric and verbal reasoning test, and reaction time.Cognitive decline was defined as an increased number of attempts on the prospective memory test, an increased number of incorrect matches on pairs matching, a decrease in the numeric and verbal reasoning test scores, or a reaction time slowed by at least 100 msec. | Age, sex, education, height, intraocular pressure, race/ethnicity, socioeconomic deprivation | Each quintile lower macular RNFL thickness (versus the most thick quintile) was significantly associated with an 18% increased risk of cognitive decline (OR = 1.18; 95% CI [1.08, 1.29]; P = 0.001). | Ko, 2018 (57); |
| N=2,456 (3-years follow-up) |
Global cognitive performance was estimated from the verbal fluency test, 15-word learning test, letter-digit substitution test, Stroop test, and Purdue pegboard test. In addition cognitive performance was estimated from the MMSE. |
Age, sex, education, BMI, blood pressure, antihypertensive medication, total cholesterol, HDL, diabetes, smoking | Lower peripapillary RNFL thickness was not significantly associated with lower global cognitive performance over time (per SD lower, -0.007; 95% CI [-0.030, 0.017] SD; P > 0.05). Lower peripapillary RNFL thickness was not significantly associated with a lower MMSE score over time (per SD lower, -0.004; 95% CI [-0.040, 0.032] SD; P > 0.05). |
Mutlu, 2018 (58); |
| N=865 (baseline measurements in childhood [age 7-11 years old] and follow -up measurement at age 45 years old) |
Change in cognitive performance from childhood (average of cognitive performance at ages 7, 9, and 11 years) to adulthood at (cognitive performance assessed at age 45 years). From the Wechsler Intelligence Scale for Children–Revised test and the Wechsler Adult Intelligence Scale test the following cognitive domains were assessed: Full Scale IQ, verbal comprehension, perceptual reasoning, and processing speed. |
Sex, intraocular pressure, axial length, and optic disc area | Lower RNFL thickness was significantly associated with a reduction in processing speed from childhood to adulthood, but not with a change in global cognitive performance (estimated with the Full Scale IQ), verbal comprehension or perceptual reasoning from childhood to adulthood. | Barrett-Young, 2022 (59); |
| Cross-sectional studies | ||||
| N=32,038 | Global cognitive performance was estimated from the following tests: prospective memory, pairs matching, numeric and verbal reasoning test, and reaction time.Test failure was defined as an incorrect answer on the first attempt of prospective memory, or doing worse than 95% of participants in pairs matching (>2 incorrect matches), numeric and verbal reasoning tests (score <3), or reaction time (>770 ms). | Age, sex, education, height, intraocular pressure, race/ethnicity, socioeconomic deprivation | The thinnest RNFL quintile (RNFL ≤ 45.9 µm) was 11% more likely to fail ≥ 1 cognitive test compared to the thickest RNFL (≥ 60.2 µm) quintile (OR = 1.11; 95% CI [1.02, 1.21]; P = 0.01). | Ko, 2018 (57); |
| N=5,347 | MMSE, Boston Naming Test, Phonemic fluency, Semantic fluency animals, Trail Making Tests, Visuoconstruction copy, Visuoconstruction delayed recall, Wordlist learning, Wordlist delayed recall and Wordlist recognition, Reading the Mind in the Eyes test, Multiple choice German vocabulary test, and the Stroop-Test. | Age, sex, radius, education | Semantic fluency, Stroop test, and the Trail Making Tests (but not other tests) were significantly associated with peripapillary RNFL thickness. | Girbardt, 2021 (60); |
| N= 5,563 | Short-form Mini-Mental State Examination, an animal naming task, a letter cancellation task, the Hopkins Verbal Learning Test, the National Adult Reading Test, and the Cambridge Neuropsychological Test Automated Battery Paired Associates Learning Test. | Disc area, age, sex, educational level, social class, visual acuity of the better eye, axial length, and a history of cataract surgery | A thicker RNFL thickness was associated with better scores for the short-form MMSE (per SD 0.06; 95% confidence interval [CI], [0.02, 0.10], P = 0.005), Hopkins Verbal Learning Test (0.16, 95% CI [0.03, 0.29]; P = 0.014), and The National Adult Reading Test (−0.24, 95% CI [−0.46, −0.02], P = 0.035). RNFL thickness was not associated with other cognitive indices. | Khawaja, 2016 (61); |
| N=2,483 | Participants completed a 50-minute cognitive test battery that assessed 5 core cognitive domains across multiple modalities. To calculate global cognitive performance results of tests for subdomains were combined. The following domains were assessed. Executive function was assessed with a verbal fluency test (total number of animals named in 60 seconds), Trail-Making Test Part B (time to completion), and an eye-tracker anti-saccade task (percentage of erroneous prosaccades). Processing speed was assessed with Trail-Making Test Part A (time to completion) and an eye-tracker prosaccade task (mean latency). Working memory was assessed with a Corsi block tapping test (sum of forward and backward span) and a digit span test (sum of forward and backward span). Verbal episodic memory was assessed with 2 components of a verbal learning and memory test (immediate recall trials 1–5 total, delayed recall). Crystallized intelligence was assessed with a multiple-choice vocabulary intelligence test (total real words correctly identified). | Age, sex, education, spherical equivalent, visual acuity, smoking, diabetes, hypertension, history of stroke | Lower peripapillary RNFL thickness was not significantly associated with a lower global cognitive performance (per SD lower RNFL thickness = -0.021; 95% CI [-0.049, 0.007] SD; P> 0.05). | Ward, 2020 (62); |
| N=3,243 | Global cognitive performance was estimated from the verbal fluency test, 15-word learning test, letter-digit substitution test, Stroop test, and Purdue pegboard test. In addition, cognitive performance was assessed with the MMSE. |
Age, sex, education, BMI, blood pressure, antihypertensive medication, total cholesterol, HDL, diabetes, smoking | Lower peripapillary RNFL thickness was significantly associated with lower global cognitive performance (per SD lower, -0.044; 95% CI [-0.074, -0.013] SD; P < 0.05). Lower peripapillary RNFL thickness was significantly associated with a lower MMSE score (per SD lower, -0.056; 95% CI [-0.090, -0.022] SD; P < 0.05). |
Mutlu, 2018 (58); |
| N=1,602 | MMSE | Age, sex, education, socio-economic status, spherical equivalent, intraocular pressure, disc area | Lower macular RNFL thickness was significantly associated with a lower MMSE score (β = NR; 95% CI [NR]; P = 0.001). | Jones-Odeh, 2016 (63); |
| N=1,485 | Different domains of cognitive functioning were assessed with the Dutch Adult Reading Test, the Rey Auditory Verbal Memory Test, semantic fluency, the Trail-Making Test, the Stroop Color-Word Test, and the Block Design test. | Age, sex, level of inbreeding, and refractive error | RNFL thickness was significantly associated with all tests of cognitive function, except for the Stroop Color-Word Test part III. | Van Koolwijk, 2009 (64); |
| N=865 | Cognitive performance was assessed as the Full Scale IQ score at age 45 years. | Sex, intraocular pressure, axial length, and optic disc area | Lower RNFL thickness was significantly associated with a lower cognitive performance. | Barrett-Young, 2022 (59); |
| N=930 (patients from a memory clinic) |
Cognitive status was assessed (N=414 cognitively healthy individuals; N=192 probable amnestic MCI; and N=324 probable AD) |
Age, education, sex, and OCT image quality | No significant differences were observed between cognitively healthy individuals, individuals with probably amnestic MCI, or individuals with probable AD. | Sanchez, 2018 (65); |
| Retinal arteriolar diameter | ||||
| Prospective studies | ||||
| N=1,134 14-year follow-up. |
Cognitive decline was defined as the 10% rapidest decline in test scores for the Delayed Word Recall Test (cut-off ≤−2.1/decade), Word Fluency Test (cut-off ≤−9.3/decade), and Digit Symbol Substitution Test (cut-off ≤−8.5/decade). | Age, sex, race, research center, hypertension, smoking, APOe4 status, diabetes | Individuals with generalized arteriolar narrowing (defined as the narrowest 25% of the population) versus individuals without generalized arteriolar narrowing did not have a greater cognitive decline based on the Delayed Word Recall Test (odds ratio [92%CI], 1.10 [0.63; 1.93]; the Word Fluency Test (1.00 [0.55; 1.80]), or the Digit Symbol Substitution Test (OR 0.95 [0.54-1.70]). | Lesage,2009 (66); |
| Cross-sectional studies | ||||
| N=1,988 | Cognitive impairment assessed with the MMSE (cutoff for impairment MMSE score<23). | Age, sex, diabetes mellitus, smoking, post‐high school education, systolic blood pressure, and history of cardiovascular disease. | In individuals with hypertension, narrow versus wide retinal arteriolar diameter (defined as lowest quintile versus highest quintile) was not significantly associated with greater cognitive impairment (beta, lowest versus highest quintile, odds ratio 2.2 [0.9–5.2]). In individuals without hypertension, narrow versus wide retinal arteriolar diameter (defined as lowest quintile versus highest quintile) was not associated with greater cognitive impairment (beta, lowest versus highest quintile, odds ratio 0.4 [0.1–1.3]). | Liew, 2009 (67); |
| N=1,744 | Executive function was assessed with the Digit symbol substitution test. | Age, sex, race, current smoking, heart disease, systolic blood pressure, natural logarithm of carotid intima-media thickness, and C-reactive protein; years of education, diabetes status, blocks walked per week, alcohol consumption; retinal venular diameter | Individuals with the narrowest retinal arteriolar diameter (narrowest 10% of the population versus the rest of the population) had a significantly lower score on the Digit symbol substitution test (beta, narrow versus wide, −1.26 [−2.39,−0.13] points). | Kim, 2011 (68); |
| N=8,600 | Verbal learning and memory assessed with the delayed Word Recall Test (cut-off for low ≤4); executive function assessed with the Digit Symbol Subtest of the Wechsler Adult Intelligence Scale–Revised (cut-off for low ≤20); verbal fluency assessed with the Word Fluency Test of the Multilingual Aphasia Examination (cut-off for low ≤11). | Age, sex, race, research center, education, occupation, diabetes, fasting glucose, hypertension and mean arterial blood pressure averaged over visits 1 through 3, carotid intima-media thickness, cigarette smoking, alcohol consumption, fasting total and high density lipid cholesterol levels, and triglyceride levels. | Individuals with generalized arteriolar narrowing (defined as the narrowest 20% of the population based on the arteriolar-to-venular ratio), versus individuals without generalized arteriolar narrowing, did not have a lower score on the delayed word recall test (odds ratio [95%CI] 1.10 [0.80; 1.49]), the digit symbol subtest of the Wechsler Adult Intelligence Scale-Revised (odds ratio 1.08 [0.69-1.69]); or the Word Fluency Test of the Multilingual Aphasia Examination (odds ratio 1.37 [0.88; 2.12]). | Wong, 2002 (69); |
| N=2,211 | Cognitive performance was assessed with the Modified Mini-Mental State Examination. Executive function was assessed with the Digit–Symbol Substitution Test. |
Age, sex, race, field center, education level, internal carotid intima-media thickness, body mass index, hypertension, diabetes status, and cigarette smoking status | Test scores for Modified Mini-Mental State Examination or the Digit–Symbol Substitution Test did not differ between different quintiles of retinal arteriolar diameter. | Baker, 2007 (70); |
| N=954 | Major domains of cognitive ability were assessed using tests of immediate and delayed non-verbal memory and verbal declarative memory (Faces and Family Pictures Subtest, Logical Memory subtest from the Wechsler Memory Scale-III); non-verbal reasoning, working memory, information processing speed (matrix reasoning, letter–number sequencing, Digit Symbol Test from the Wechsler Adult Intelligence Scale 3rd Edition); executive function (Borkowski Verbal Fluency Test); and mental flexibility (Trail Making Test-Part B). In addition, Vocabulary (‘crystallized’ intelligence) was measured (using the combined version of the Junior and Senior Mill Hill Vocabulary Scale synonyms) and global cognitive performance was assessed with the MMSE. |
Age, sex, level of education, smoking, systolic blood pressure, major macrovascular disease and depression | Wider retinal arteriolar diameter was significantly associated with a lower memory test score in men, but not in women. In addition, there were no significant associations of retinal arteriolar diameter with other cognitive tests under study. | Ding, 2011 (71); |
| N=809 | Cognitive screening was conducted during the in-home interview with the CASI-S, an abbreviated version of the Cognitive Abilities Screening Instrument The CASI-S includes four subtests: verbal registration (immediate recall of three words), temporal orientation, verbal fluency and verbal recall (delayed recall of three words). In a subset (n=281) cognitive performance was assessed with the SENAS test, which consists of the following measures: word list learning, spatial configuration learning, category fluency (animals & supermarket), phonemic fluency (letters F & L), list sorting (two lists), digit span forward, digit span backward, verbal conceptual thinking, object naming, picture association, pattern recognition and spatial localization that assess both verbal and non-verbal aspects of cognition |
Age, sex, educational level, language of exam administration (Spanish, English), smoking, hypertension | Narrower retinal arteriolar diameters (continuous variable) and generalized arteriolar narrowing (defined as the lowest 25% of the data versus the rest of the data) were not statistically significantly associated with lower cognitive performance (estimated from the CASI-S test and the SENAS test). | Gatto, 2012 (72); |
| Retinal venular diameter | ||||
| Prospective studies | ||||
| No studies available | ||||
| Cross-sectional studies | ||||
| N=1,988 | Cognitive impairment assessed with the MMSE (cutoff for impairment MMSE score<23). | Age, sex, diabetes mellitus, smoking, post‐high school education, systolic blood pressure, and history of cardiovascular disease. | In individuals with hypertension, wide versus narrow retinal venular diameter [defined as highest quartile of data versus other quartiles] was significantly associated with greater cognitive impairment (odds ratio 2.7 [1.2–6.1]). Alternatively, in individuals without hypertension, wide versus narrow retinal venular diameter was not associated with greater cognitive impairment (odds ratio 1.0 [0.4–2.4]). | Liew, 2009 (67); |
| N=1,744 | Executive function was assessed with the Digit symbol substitution test. | Age, sex, race, current smoking, heart disease systolic blood pressure, natural logarithm of carotid intima-media thickness, and C-reactive protein; years of education, diabetes status, blocks walked per week, alcohol consumption; retinal arteriolar diameter |
Individuals with the widest retinal venular diameter (widest 10% of the population versus the rest of the population) did not have a significantly lower score on the Digit symbol substitution test (1.35 [−0.12, 2.82]). | Kim, 2011 (68); |
| N=2,211 | Cognitive function assessed with the Modified Mini-Mental State Examination Executive function was assessed with the Digit–Symbol Substitution Test. |
Age, sex, race, field center, education level, internal carotid intima-media thickness, body mass index, hypertension, diabetes status, and cigarette smoking status. | Test scores for the Modified Mini-Mental State Examination or the Digit–Symbol Substitution Test did not differ between different quintiles of retinal venular diameter. | Baker, 2007 (70); |
| N=954 | Major domains of cognitive ability were assessed using tests of immediate and delayed non-verbal memory and verbal declarative memory (Faces and Family Pictures Subtest, Logical Memory subtest from the Wechsler Memory Scale-III); non-verbal reasoning, working memory, information processing speed (matrix reasoning, letter–number sequencing, Digit Symbol Test from the Wechsler Adult Intelligence Scale 3rd Edition); executive function (Borkowski Verbal Fluency Test); and mental flexibility (Trail Making Test-Part B). In addition, Vocabulary (‘crystallized’ intelligence) was measured (using the combined version of the Junior and Senior Mill Hill Vocabulary Scale synonyms) and global cognitive performance was assessed with the MMSE. |
Age, sex, level of education, smoking, systolic blood pressure, major macrovascular disease and depression | Wider retinal venular diameter was significantly associated with a lower memory test score in men, but not in women. In addition, there were no significant associations of retinal arteriolar diameter with other cognitive tests under study. | Ding, 2011 (71); |
| N=809 | Cognitive screening was conducted during the in-home interview with the CASI-S, an abbreviated version of the Cognitive Abilities Screening Instrument The CASI-S includes four subtests: verbal registration (immediate recall of three words), temporal orientation, verbal fluency and verbal recall (delayed recall of three words). In a subset (n=281) cognitive performance was assessed with the SENAS test, which consists of the following measures: word list learning, spatial configuration learning, category fluency (animals & supermarket), phonemic fluency (letters F & L), list sorting (two lists), digit span forward, digit span backward, verbal conceptual thinking, object naming, picture association, pattern recognition and spatial localization that assess both verbal and non-verbal aspects of cognition |
Age, sex, educational level, language of exam administration (Spanish, English), smoking, hypertension | A non-significant trend between generalized venular widening (defined as the highest 25% of the data versus the rest of the data) and lower cognitive performance was found (for both the CASI-S and the SENAS tests). However, wider retinal venular diameters, when analyzed as continuous measures, were not statistically significantly associated with lower cognitive performance (estimated from the CASI-S and SENAS tests). | Gatto, 2012 (72); |
| Flicker light-induced increase in retinal microvascular diameters | ||||
| Prospective studies | ||||
| No studies available | ||||
| Cross-sectional studies | ||||
| N=2,018 (arteriole); And N=2,049 (venule); Community-dwelling participants; oversampling of individuals with type 2 diabetes. |
Global cognitive performance was calculated as the mean cognitive performance for the domains memory, information processing, and executive function. The composite memory score was derived from the Verbal Learning Test by weighting total immediate and delayed recall scores. The domain information processing speed included the Stroop Color-Word Test Part I and II, the Concept Shifting Test Part A and B, and the Letter Digit Substitution Test. Executive function was assessed by the Stroop Color-Word Test Part III and the Concept Shifting Test Part C. | age, education level, sex, type 2 diabetes, body mass index, smoking status, alcohol use, hypertension, total/high density lipid cholesterol ratio, triglycerides, lipid-modifying medication use, prior cardiovascular disease, current depression and plasma biomarkers of low-grade inflammation | Lower percentage flicker light-induced increase in retinal arteriolar or venular diameter was not significantly associated with lower cognitive performance (per SD lower arteriolar and venular percentage flicker light-induced increase [in SD], respectively, -0.016 [-0.050; 0.019]; and -0.022 [-0.056; 0.011]). | Rensma, 2019 (73); |
| N=56; N=12 individuals with AD; n= 12 individuals with MCI; and n=32 sex- and age-matched controls. | Diagnosis of AD and MCI was based on the National Institute on Aging-Alzheimer’s Association criteria. | None | The mean flicker light-induced percentage increase in retinal arteriolar diameter was significantly lower in individuals with AD (mean percentage increase: 0.21 ± 1.80%) and in individuals with MCI (mean percentage increase: 2.29 ± 1.81%) than in healthy controls (mean percentage increase: 3.86 ± 1.94%). The mean flicker light-induced percentage increase in retinal venular diameter was significantly lower in individuals with AD (2.84 ± 1.56; P<0.05) than in individuals with MCI (3.45 ± 1.96) or healthy controls (3.37 ± 1.81), however, there were no significant differences between individuals with MCI or healthy controls. |
Querques, 2019 (74); |
| N=54; N=15 individuals with AD; n= 24 individuals with MCI; and n=15 healthy controls. | Diagnosis of AD and MCI was based on the National Institute on Aging-Alzheimer’s Association criteria. | None | The percentage mean maximal flicker light-induced increase in retinal arteriolar diameter was significantly greater in individuals with AD in comparison with healthy controls (P = 0.004), and numerically in between in individuals with MCI (though not statistically significantly differently from healthy controls). The percentage mean maximal flicker light-induced increase in retinal venular diameter was significantly greater in individuals with AD in comparison with healthy controls (P = 0.001), and numerically in between (though not statistically significant from healthy controls) in individuals with MCI. |
Kotliar, 2017 (75); |
| Retinal blood flow | ||||
| Prospective studies | ||||
| No studies available | ||||
| Cross-sectional studies | ||||
| N=52; N=10 individuals with AD; n= 21 individuals with MCI; and n=21 healthy controls. | MCI diagnosis was based on the 2004 MCI Working Group Criteria for amnestic MCI and on the global clinical dementia rating. AD diagnosis was based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) diagnostic criteria for dementia and National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria. |
None | In individuals with AD (mean venous blood speed 29.1 ± 6.3 mm/s) and MCI (mean venous blood speed 28.9 ± 6.3 mm/s) the venous blood speed was significantly lower than in healthy control individuals (mean venous blood speed 36.0 ± 9.3 mm/s) In individuals with AD (mean venous blood flow 12.3 ± 2.8 μl/min) and MCI (mean venous blood flow 16.1 ± 4.4 μl/min) the venous blood flow was significantly lower than in healthy control individuals (mean venous blood flow 20.0 ± 5.4 μl/min). |
Feke, 2015 (76); |
| N=17; N=9 individuals with AD; and n=8 healthy controls. | AD diagnosis was based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) diagnostic criteria for dementia and National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria. |
None | The venous blood flow rate was significantly lower in individuals with AD (9.7 ± 3.1 μL/min) than in healthy controls (15.9 ± 3.7 μL/min, P = 0.002). The venous centerline blood speed was lower, but not statistically significantly, in individuals with AD (23.8 ± 8.2 mm/s) than in healthy controls (30.8 ± 6.0 mm/s; P = 0.07). |
Berisha, 2007 (77); |
| Vascular density and size of the foveal avascular area | ||||
| Prospective studies | ||||
| N=297; 10-years follow-up; Community-dwelling participants |
A 10-test neurocognitive battery spanning several cognitive domains (memory and speed of processing/executive function) was administered. Diagnosis of AD and MCI was based on the National Institute on Aging-Alzheimer’s Association criteria. |
age, income, education, race/community (black/Jackson or white/Washington Co), current smoking, physical activity, hypertension, and diabetes. | Greater superficial vascular complex density, greater deep capillary plexus vessel density and greater foveal avascular zone area were not significantly associated with change in global cognitive performance (per 10% increase in superficial vascular complex density or deep capillary plexus vessel density or per 1 mm2 larger foveal avascular zone area, respectively, 0.04 [−0.07, 0.15]; 0.03 [−0.08, 0.14], and 0.24 [−0.35, 0.83] SD). | Abraham, 2021 (78); |
| Cross-sectional studies | ||||
| N=571 (Meta-analysis; N=150 individuals with AD; n= 195 individuals with MCI; and n=226 healthy controls |
AD and MCI were defined in individual studies. | None | Individuals with MCI, versus healthy controls, had a significantly lower vessel density in the superficial capillary plexus: weighted mean difference: -2.26, 95% CI: -3.98 to -0.55, P = 0.01; in the deep capillary plexus: weighted mean difference: -3.40, 95% CI: -5.99 to -0.81, P = 0.01; and a larger fovea avascular zone area: weighted mean difference = 0.06, 95% CI: 0.01 to 0.11, P = 0.02. Individuals with AD, versus healthy controls, had a significantly lower vessel density in the superficial capillary plexus: weighted mean difference = -1.88, 95% CI: -2.7 to -1.07, p<0.00001. In contrast, there was no significant decrease in the deep capillary plexus nor enlargement of the avascular zone in individuals with AD. |
Hui, 2021 (79); |
| N=297; Community-dwelling participants |
A 10-test neurocognitive battery spanning several cognitive domains (memory and speed of processing/executive function) was administered. Diagnosis of AD and MCI was based on the National Institute on Aging-Alzheimer’s Association criteria. |
Age, sex, income, education, race/community (black/Jackson or white/Washington Co), current smoking, physical activity, hypertension, and diabetes | Greater superficial vascular complex density, greater deep capillary plexus vessel density and greater foveal avascular zone area were not associated with global cognitive performance (per 10% increase in superficial vascular complex density or deep capillary plexus vessel density or per 1 mm2 larger foveal avascular zone area, respectively, −0.02 [−0.7, 0.01]; −0.01 [−0.06, 0.09] and 0.23 [−0.31, 0.77] SD). | Abraham, 2021 (78); |
| N=177; Chronic kidney disease stage ≥3 (including end-stage renal disease). | MMSE | The following variables were combined in 1 model: RNFL thickness, ganglion complex cell thickness, superficial vascular complex density, and deep capillary plexus vessel density, age, sex, estimated glomerular filtration rate, diabetes mellitus, hypertension, dyslipidemia, and years of education | Greater superficial vascular complex density was not associated with global cognitive performance but greater deep capillary plexus vessel density was (per percentage greater superficial vascular complex density and deep capillary plexus vessel density, respectively, -0.07 [-0.16; 0.01] and 0.11 [0.01; 0.21]). | Peng, 2021 (80); |
Table 3 shows the associations of neuronal and microvascular structures and function of the neurovascular coupling unit, as quantified by retinal imaging, with cognitive performance. Fully adjusted results are shown (where appropriate, i.e. when associations where adjusted for confounders).
MMSE, Mini-Mental State Examination; RNFL, retinal nerve fiber layer thickness; OR, odds ratio; SD, standard deviation; CI, confidence interval; MCI, mild cognitive impairment; AD, Alzheimer’s disease.