Exploration of Rapid Automatized Naming and Standard Visual Tests in Prodromal Alzheimer Disease Detection

Visual tests in Alzheimer disease (AD) have been examined over the last several decades to identify a sensitive and noninvasive marker of the disease. Rapid automatized naming (RAN) tasks have shown promise for detecting prodromal AD or mild cognitive impairment (MCI). The purpose of this investigation was to determine the capacity for new rapid image and number naming tests and other measures of visual pathway structure and function to distinguish individuals with MCI due to AD from those with normal aging and cognition. The relation of these tests to vision-specific quality of life scores was also examined in this pilot study.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Exploration of Rapid Automatized Naming and Standard Visual Tests in Prodromal Alzheimer Disease Detection Shirley Z. Wu, BS, Rachel Nolan-Kenney, MPhil, Nicholas J. Moehringer, BS, Lisena F. Hasanaj, MPA, Binu M. Joseph, MBBS, Ashley M. Clayton, MA, Janet C. Rucker, MD, Steven L. Galetta, MD, Thomas M. Wisniewski, MD, Arjun V. Masurkar, MD, PhD, Laura J. Balcer, MD, MSCE Background: Visual tests in Alzheimer disease (AD) have been examined over the last several decades to identify a sensitive and noninvasive marker of the disease. Rapid automatized naming (RAN) tasks have shown promise for detecting prodromal AD or mild cognitive impairment (MCI). The purpose of this investigation was to determine the capacity for new rapid image and number naming tests and Departments of Neurology (SZW, RNK, NM, LH, BJ, AC, JCR, SLG, TMW, AVM, and LJB), Population Health (RNK and LJB), and Ophthalmology (SZW, JCR, SLG, and LJB), New York University Grossman School of Medicine, New York, New York. Supported by the National Institutes of Health research grants P30AG008051, P30AG066512, and UL1TR001445. J. C. Rucker is on the Editorial Board of the Journal of NeuroOphthalmology. S. L. Galetta has been a consultant to Biogen. A. V. Masurkar is on the Council of the Alzheimer’s Association International Research Grants Program, the Steering Committee of the Alzheimer’s Disease Cooperative Study, and on the Advisory Board of the Journal of Neuro-Ophthalmology. L. J. Balcer is Editor-in-Chief of the Journal of Neuro-Ophthalmology. The remaining authors report no conflicts of interest. Author Contributions: S. Z. Wu: conception and design, acquisition of data, data analysis, interpretation of data, drafting the manuscript, revising for intellectual content, and final approval of the completed article. R. Nolan-Kenney: conception and design, acquisition of data, data analysis, interpretation of data, drafting the article, revising for intellectual content, and final approval of the completed article. N. Moehringer: acquisition of data and final approval of the completed article. L. Hasanaj: acquisition of data and final approval of the completed article. B. Joseph: acquisition of data and final approval of the completed article. A. Clayton: acquisition of data and final approval of the completed article. J. C. Rucker: interpretation of data, revising for intellectual content, and final approval of the completed article. S. L. Galetta: interpretation of data, revising for intellectual content, and final approval of the completed article. T. M. Wisniewski: interpretation of data, revising for intellectual content, and final approval of the completed article. A. V. Masurkar: conception and design, interpretation of data, revising for intellectual content, and final approval of the completed article. L. J. Balcer: conception and design, interpretation of data, revising for intellectual content, and final approval of the completed article. Address correspondence to Laura J. Balcer, MD, MSCE, Department of Neurology, NYU Grossman School of Medicine, 222 East 41st Street, New York, NY 10017; E-mail: [e-mail] Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 other measures of visual pathway structure and function to distinguish individuals with MCI due to AD from those with normal aging and cognition. The relation of these tests to vision-specific quality of life scores was also examined in this pilot study. Methods: Participants with MCI due to AD and controls from well-characterized NYU research and clinical cohorts performed high and low-contrast letter acuity (LCLA) testing, as well as RAN using the Mobile Universal Lexicon Evaluation System (MULES) and Staggered Uneven Number test, and vision-specific quality of life scales, including the 25-Item National Eye Institute Visual Function Questionnaire (NEI-VFQ-25) and 10-Item NeuroOphthalmic Supplement. Individuals also underwent optical coherence tomography scans to assess peripapillary retinal nerve fiber layer and ganglion cell/inner plexiform layer thicknesses. Hippocampal atrophy on brain MRI was also determined from the participants’ Alzheimer disease research center or clinical data. Results: Participants with MCI (n = 14) had worse binocular LCLA at 1.25% contrast compared with controls (P = 0.009) and longer (worse) MULES test times (P = 0.006) with more errors in naming images (P = 0.009) compared with controls (n = 16). These were the only significantly different visual tests between groups. MULES test times (area under the receiver operating characteristic curve [AUC] = 0.79), MULES errors (AUC = 0.78), and binocular 1.25% LCLA (AUC = 0.78) showed good diagnostic accuracy for distinguishing MCI from controls. A combination of the MULES score and 1.25% LCLA demonstrated the greatest capacity to distinguish (AUC = 0.87). These visual measures were better predictors of MCI vs control status than the presence of hippocampal atrophy on brain MRI in this cohort. A greater number of MULES test errors (r s = 20.50, P = 0.005) and worse 1.25% LCLA scores (r s = 0.39, P = 0.03) were associated with lower (worse) NEI-VFQ-25 scores. Conclusions: Rapid image naming (MULES) and LCLA are able to distinguish MCI due to AD from normal aging and reflect vision-specific quality of life. Larger studies will determine how these easily administered tests may 79 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution identify patients at risk for AD and serve as measures in disease-modifying therapy clinical trials. Journal of Neuro-Ophthalmology 2022;42:79–87 doi: 10.1097/WNO.0000000000001228 © 2021 by North American Neuro-Ophthalmology Society INTRODUCTION B iomarkers of early Alzheimer disease (AD) play a critical role in clinical trials of disease-modifying therapies by identifying appropriate research participants and monitoring efficacy of treatment. Biomarkers are also important because they assist in identifying at-risk individuals who would benefit from anticipatory guidance and any future treatments. Widely accepted AD biomarkers based on cerebrospinal fluid analysis and imaging studies require expensive and invasive procedures that are not routinely performed (1). The search for sensitive, practical, and noninvasive markers of disease makes visual measures an area of interest. In recent decades, visual biomarkers have been examined in AD. Such studies were driven by visual changes in early disease and by evidence of AD pathology in vision-related structures of the brain and eye (2–4). Studies have found that patients with AD dementia have significantly impaired contrast sensitivity, color vision, visual attention, and saccadic eye movements compared with controls (5–7). Thinning of the peripapillary retinal nerve fiber layer (pRNFL) and macula by optical coherence tomography (OCT) has also been observed in AD dementia patients compared with controls (8). Fewer studies evaluate these visual measures in prodromal AD or mild cognitive impairment (MCI). For some visual measures, it remains controversial whether these metrics have the capacity to differentiate patients with disease from controls at this stage (8,9). Therefore, there is a need to evaluate the performance of new visual tests and reassess previously studied visual measures in MCI cohorts. In addition to disease detection, it is important to evaluate how visual measures correlate with vision-specific quality of life (QOL), a measure of patient experience, function, and well-being. More studies are needed to explore this topic, as vision-specific QOL could potentially also inform the disease status. Identifying visual changes that decrease QOL would also inform caretakers on type of accommodations to best assist patients. Rapid automatized naming (RAN) tests, although relatively unexplored in early AD, are vision-based tests that have been shown to detect concussion, multiple sclerosis, and other neurodegenerative diseases such as Parkinson disease (10–12). RAN tests are timed and require quick and accurate identification and naming aloud of 80 numbers or images. Accurate performance requires intact brain pathways involved in visual perception, color vision, object categorization, object-specific memory, eye movements, and attention. These functions may be impaired in AD, increasing the potential sensitivity of these tests for detecting early stage disease. One study to date has investigated RAN test performance in AD and found that rapid number naming, in the form of the King–Devick test, differentiated AD dementia and MCI from controls (13). Our group has developed 2 RAN tests, neither of which have been studied in AD. The Mobile Universal Lexicon Evaluation System (MULES) is a rapid image naming test composed of fruits, animals, and objects in context (10–12), and the Staggered Uneven Number (SUN) test is a newly developed rapid number naming test composed of numbers in zigzag and horizontal patterns with unique spacing (14). The purpose of this investigation was to determine the capacity for new rapid image and number naming tests and other measures of visual pathway structure and function to distinguish patients with MCI due to AD from those with normal aging and cognition. The relation of these tests to vision-specific QOL scores was also examined in this pilot study. METHODS Study Participants A convenience sample of participants who were 65 years of age or older, English-speaking, and had a diagnosis of MCI due to AD or normal cognition (control participants) was recruited from the NYU Alzheimer’s Disease Research Center and the Pearl I. Barlow Center for Memory Evaluation and Treatment. The diagnosis of MCI, with intermediate to high likelihood of AD as the etiology, was based on National Institute of Aging- Alzheimer’s Association (NIA-AA) criteria (15). Diagnostic criteria included a Global Clinical Dementia Rating (CDR) score of 0.5, abnormal psychometric testing with amnestic pattern, and established AD biomarker assessment (neurodegeneration assessed by MRI or FDG PETMRI, with a subset also having CSF amyloid/tau analysis or amyloid PET scan). Controls had a CDR score of 0 and no abnormalities on psychometric testing. Participants with severe psychiatric or neurological disease, including other types of dementias, were excluded. Individuals with comorbid ocular disease were not excluded to allow for more accurate representation of the population affected by Alzheimer disease, but ocular pathology was later accounted for in statistical analyses. All participants received a fifty-dollar gift card for their participation in the study. The study protocol, including study compensation, was approved by the NYU Grossman School of Medicine Institutional Review Board; all participants provided written informed consent. Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Vision Testing Statistical Analysis Vision testing was assessed monocularly and binocularly in controlled lighting with participants’ best distance correction. High-contrast or best-corrected visual acuity (BCVA) was evaluated with retro-illuminated Early Treatment Diabetic Retinopathy Study charts at a 3.2-m distance. The low-contrast letter acuity (LCLA) was determined using retro-illuminated lowcontrast Sloan letter charts (Precision Vision, La Salle, IL) at both 2.5% and 1.25% contrast levels at a 2-m distance. The number of letters identified correctly out of a maximum of 70 was recorded. Statistical analyses were performed using SPSS 26.0 (SPSS, Inc, Chicago, IL). Before any analyses, OCT scans with insufficient quality or from eyes with reported or incidental retinal or optic nerve pathology were excluded from corresponding OCT analyses. Eyes with significant visual impairment (e.g., amblyopia) were excluded from monocular vision analyses. Wilcoxon rank-sum or independent (two-sample) t-tests, depending on how well distributions of continuous variables met assumptions for normality, were used to identify visual tests that significantly differed between MCI participants and controls. Logistic regression analyses were performed to calculate unadjusted and adjusted odds ratios for MCI vs control status based on vision test scores, accounting simultaneously for ocular pathology, age, gender, education, or best-corrected visual acuity. To account for these factors, they were individually adjusted for in separate logistic regression models. Receiver operating characteristic (ROC) curves were generated to determine the capacity of the visual tests to distinguish between participants with MCI and controls; high-sensitivity score cut-offs based on the ROC analyses were determined. In a secondary analysis, the presence or absence of MRI hippocampal atrophy, determined by a cognitive neurologist masked to MCI vs control status, was examined for capacity to distinguish MCI vs controls. It is important to note that this assessment of hippocampal atrophy differs from the imaging criteria used to support a diagnosis of MCI due to AD. Neurodegeneration as part of the diagnostic criteria was based on the global cortical signature of AD (temporoparietal atrophy on MRI or hypometabolism on FDG-PET), whereas for this analysis, the hippocampal formation was qualitatively assessed in isolation. To determine the relation of visionspecific QOL with visual measures, Spearman rank correlation coefficients were calculated. Cohen’s d guidelines were used to estimate the magnitude of correlations with 0.1 considered a small effect, 0.3 a medium effect, and 0.5 a large effect (21). Rapid Automatized Naming The MULES is a double-sided laminated 8.5 · 11-inch paper with 54 color photographs of fruits, objects, and animals in context (with their usual backgrounds) (Fig. 1A) (10–12). The SUN test is a single-sided 8.5 · 11- inch paper with 145 single- and double-digit numbers arranged in zigzag or horizontal patterns with varied spacing (Fig. 1B) (14). Participants were asked to name aloud each image or number as quickly and accurately as possible from left to right in the top row and every subsequent row. Two trials were performed, and the total time in seconds to complete the test and number of errors for each trial were recorded for the MULES and SUN. Inability to identify, misidentifying, or skipping a image or number counted as an error. Optical Coherence Tomography Participants underwent optic nerve head and macula scans of both eyes using Cirrus and Spectralis spectral-domain OCT (Carl Zeiss Meditec, Dublin, CA; Heidelberg Engineering, Heidelberg, Germany). The Spectralis OCT device was used because it is more sensitive than Cirrus OCT at detecting retinal atrophy, especially in an Alzheimer disease population (16). Cirrus OCT was also included because, unlike Spectralis, its software automatically isolates and generates the thickness of the ganglion cell/inner plexiform layer (GCIPL). Using both devices, macular volume and pRNFL, macular, and GCIPL thicknesses were measured. Specific OCT protocols and methods used including quality control of scans are previously published (17). Vision-specific Quality of Life Questionnaires Vision-specific QOL was measured by the 25-Item National Eye Institute Visual Function Questionnaire (NEI-VFQ-25) (18), which assesses self-reported visual impairment in everyday activities. The 10-Item NeuroOphthalmic Supplement to the NEI-VFQ-25 (19), which addresses specific neuro-ophthalmic visual symptoms, was also administered. Participants were asked to answer questions as if they were wearing their glasses or contact lenses for the activity specified. Questionnaires were scored using published algorithms (20); scale scores range from 0 (worst QOL) to 100 (best QOL). Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 RESULTS Among 14 participants with MCI and 16 controls without cognitive impairment, there were no significant differences in demographic factors (Table 1). Both groups, on average, were aged 75 years, had 17 years of education, and had bestcorrected binocular visual acuities between 20/20 and 20/ 25. Qualitatively, the MCI group had more reported ocular history and fewer female participants compared with controls (Table 1). None reported significant comorbid neurologic disease. Among all studied visual measures, binocular LCLA at 1.25% contrast and MULES performance were the only measures that were significantly worse in MCI participants compared with controls (Table 2). MCI participants 81 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 82 Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. The MULES and SUN tests (A) the Mobile Universal Lexicon Evaluation System (MULES) is a rapid image naming test (MULES test New York University, text and photographs, registration number TXu002026665, all rights reserved). It is a double-sided laminated 8.5 · 11-inch paper with 54 color photographs of fruits, objects, and animals in context. B. The SUN test, a rapid number naming test, is a single-sided 8.5 · 11-inch paper with 145 single- and double-digit numbers arranged in zigzag or horizontal patterns with varied spacing (SUN test 2019 New York University. All Rights reserved). Participants are asked to read out loud each image or number as quickly and accurately as possible from left to right in the top row and every subsequent row. identified only one-third of the number of letters seen by controls at 1.25% contrast; this corresponded to a difference of 10 letters in binocular LCLA. MCI participants, on average, had MULES test times that were 1.6x longer (worse) compared with controls (difference of 35 seconds; P = 0.006) and committed 3 times more errors (P = 0.009). SUN test times and numbers of errors did not significantly differ between groups. Learning effects (improvement of time scores) between first and second trials of the MULES and SUN were not significantly different between groups; however, MULES learning effects were slightly less in the MCI group (Table 2). For OCT, we were able to image 16 control and 14 MCI participants with the Cirrus and 14 control and 11 MCI with the Spectralis. Scans from 7 controls and 10 MCI eyes were excluded from macular analysis, and scans from 4 controls and 9 MCI eyes were excluded from optic nerve head analysis. These scans were excluded because of existing or incidental ocular pathology (epiretinal membrane, macular hole, glaucoma, amblyopia, and scleral crescents) or insufficient quality. From eyes that had good quality images, pRNFL and GCIPL measurements were generally thinner among the MCI group compared with controls; these differences did not reach statistical significance (Table 2). Lower (worse) binocular LCLA scores at 1.25% contrast, longer (worse) MULES test times, and greater numbers of MULES errors were significant predictors of MCI vs control status, accounting simultaneously for ocular pathology (Table 3). These associations remained significant in models accounting for age, gender, education, and BCVA. Subjective hippocampal atrophy on brain MRI was not a significant predictor of MCI vs control status accounting for age in this cohort (Table 3). ROC curve analyses demonstrated that 1.25% LCLA, best MULES test time (between 2 trials), and best MULES error score had comparable AUCs of 0.78–79 (Fig. 2). LCLA scores at 1.25% that are less than 20–23 letters were associated with a sensitivity of .0.90 for detecting MCI, with a cut-off of 20 letters yielding a sensitivity of 0.93 and specificity of 0.38. Best MULES times of 49.3–49.7 seconds were associated with a sensitivity .0.90, with a 56.6-second cut-off yielding a sensitivity of 0.86 and specificity of 0.63. Further analyses with best MULES time and 1.25% LCLA combined had the greatest AUC with a value of 0.87 (P = 0.001). Best MULES error score and 1.25% LCLA combined had an AUC of 0.84 (P = 0.002); a combination of all 3 measures yielded an AUC of 0.86 (P = 0.001). Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 MCI participants had reduced NEI-VFQ-25 composite (overall) scores (these scores exclude the single item for general health) (Table 4). Differences in scores for the 10Item Neuro-Ophthalmic Supplement, which assesses primarily efferent aspects of vision, were not significant. Among all participants, NEI-VFQ-25 scores were lower (worse) among those with lower (worse) 1.25% LCLA scores (Table 5). Rank correlation of MULES errors with NEI-VFQ-25 scores were also significant with a large Cohen’s d effect size. DISCUSSION Results of this pilot study demonstrate that the MULES test of rapid image naming and binocular LCLA at 1.25% contrast distinguish patients with MCI due to AD from similar-aged controls without cognitive impairment. Because MCI is prodromal to AD, it is important that these visual measures are showing early promise as future markers to identify disease states and to potentially follow patients in clinical trials and observational research. Notably, both MULES and LCLA were stronger predictors of the MCI disease status than was hippocampal atrophy on MRI; however, our cohort size may have limited our findings in this regard. Our research on markers of visual TABLE 1. Demographics and clinical characteristics of MCI participants and controls N Age, yr (mean ± SD) Gender, n (%) female Education, yrs (mean ± SD) Ophthalmologic history, n (%) Cataract extraction Cataracts Glaucoma Retinal disease Others Incidental retinal finding Incidental optic nerve finding BCVA* (mean ± SD) Control MCI 16 75.4 ± 5.1 12 (75.0) 17.6 ± 1.4 11 (68.8) 6 5 1 2 0 3 1 54 ± 7 14 75.8 ± 7.1 8 (57.1) 16.6 ± 2.8 13 (92.9) 6 3 2 0 2 3 1 52 ± 5 P† 0.85 0.30‡ 0.50§ 0.43‡ 0.41 *Letters identified correctly with binocular vision (both eyes together) on Early Treatment Diabetic Retinopathy Study charts at 3.2 m, with 55 letters representing a Snellen equivalent of 20/20. † Independent samples (two-sample) t test unless otherwise indicated. ‡ 2 x analysis. § Wilcoxon rank-sum test. 83 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Differences in vision test scores between MCI participants and controls Visual Measures LCLA LCLA 2.5% OU LCLA 1.25% OU RAN Best MULES trial time (s) Best MULES trial errors Best SUN trial time (s) Best SUN trial errors MULES learning effect SUN learning effect OCT retinal thickness (mm) Average RNFL OD‡ Average RNFL OS‡ Macular volume OD‡ Macular volume OS‡ GCIPL OD§ GCIPL OS§ Group (n) Mean ± SD Control (16) MCI (14) Control (16) MCI (14) 38.8 ± 7.6 33.0 ± 11.2 15.8 ± 9.2 5.7 ± 9.0 0.10 Control (16) MCI (14) Control (16) MCI (14) Control (16) MCI (14) Control (16) MCI (14) Control (16) MCI (14) Control (16) MCI (14) 58.6 ± 16.6 93.5 ± 44.0 1.9 ± 1.9 5.9 ± 6.0 57.6 ± 12.4 61.1 ± 16.4 1.1 ± 2.7 0.9 ± 1.3 14.8 ± 7.2 9.1 ± 15.5 3.5 ± 5.3 4.4 ± 7.3 0.006** Control (12) MCI (9) Control (12) MCI (8) Control (12) MCI (9) Control (11) MCI (8) Control (13) MCI (8) Control (12) MCI (9) 95.8 ± 16.6 92.8 ± 8.1 95.8 ± 14.7 90.0 ± 9.8 8.4 ± 0.5 8.3 ± 0.4 8.4 ± 0.5 8.3 ± 0.5 78.5 ± 9.5 75.0 ± 4.6 77.2 ± 7.8 74.6 ± 3.5 0.25 P* 0.009** 0.009** 0.51† 0.76 0.22† 0.69† 0.34 0.70 0.49 0.19 0.22 **P , 0.01. *Wilcoxon rank-sum test unless otherwise indicated. † Independent samples (2-sample) t test. ‡ Values from Spectralis OCT. § Values from Cirrus OCT. OU, oculus uterque; right eye, oculus dexter; left eye, oculus sinister. pathway structure and function in MCI and AD will continue, harnessing the well-characterized nature of the NIH-funded NYU Alzheimer’s Disease Research Center cohort. Our preliminary finding of a lack of association of hippocampal atrophy with reduced visual function and retinal thinning is important because biomarkers of neuro- nal injury, including hippocampal volume loss, are included in official guidelines as methods to increase diagnostic certainty of AD etiology and provide information about disease stage (15). Our results provide early support for the addition of vision-based biomarkers to these guidelines, which may further increase diagnostic sensitivity when paired with existing biomarkers. Reduced 1.25% LCLA TABLE 3. Unadjusted and adjusted odds ratios from logistic regression models using visual or imaging measures to predict MCI vs control status Characteristic LCLA 1.25% OU (no. of correctly identified letters) Best MULES time (s) Best MULES errors MRI hippocampal atrophy (Y/N) Unadjusted OR (95% CI) 0.89 1.05 1.48 2.20 (0.81–0.97) (1.01–1.10) (1.04–2.11) (0.50–9.75) P 0.012 0.024 0.029 0.299 Adjusted OR† (95% CI) 0.89 1.07 1.50 2.26 (0.81–0.98) (1.01–1.14) (1.05–2.14) ‡ (0.47–10.80) P 0.01* 0.02* 0.03* 0.31 *P , 0.05. † Logistic regression controlling for ocular pathology unless otherwise indicated. ‡ Controlling for age. OU, oculus uterque. 84 Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Receiver operating characteristic curves demonstrating the capacities for the MULES and LCLA to distinguish participants with MCI vs controls without cognitive impairment. scores and greater numbers of MULES errors were also associated with decreased vision-specific QOL, consistent with prior findings that impaired contrast sensitivity was highly correlated with perceived visual disability (22–24). Performance on the SUN, a new rapid number naming test, did not differ between the MCI and control groups; this is in contrast to a study using the King–Devick test of rapid number naming, where AD dementia and MCI subjects had significantly worse time scores compared with controls (13). However, unlike our cohort, their MCI cohort was not specific to AD and included nonamnestic MCI subtypes, which can be prodromal stages of Lewy body dementia, frontotemporal dementia, and vascular dementia (25). The sample size was also substantially larger in that study. Dementia may also impair cognitive and vision-related pathways differently in early disease, which may, in part, explain the discrepancy between studies. Larger investigations of the SUN are needed, particularly as this test has a smaller learning effect (thus potentially more reliable longitudinally) and likely involves some different brain pathways from those related to MULES performance. RAN tasks for images vs numbers likely involve recruitment of different brain pathways for vision and other dimensions of function, such as language and concentration (26). The perirhinal cortex (PRC), involved in feature integration, assists in image recognition by discriminating the presented object from others with similar features (27,28). The PRC contains the transenterorhinal cortex, the earliest site of tau deposition (29), and atrophy and tau deposition within this structure has been found in MCI and early AD (30,31). Pathology in the PRC may occur earlier than in areas implicated in number naming such as the precuneus and bilateral supramarginal gyrus (32), leading to impaired image naming but not number naming in early disease. Functional MRI and PET studies are needed to explore which brain regions are implicated in image and numberbased RAN tests and affected by AD pathology. In addition, MULES images are more difficult to identify than SUN numbers because the MULES requires identification of the presented image from limitless images stored in memory, whereas for the SUN, there are limited numbers to choose from in stored memory (single- and some doubledigit numbers). Identification of a MULES image vs SUN number also requires additional pathways involved in object categorization, object-specific memory, and color vision; thus, subtle deficits in any of these areas may be apparent on performance on a timed image naming test. The capacity for retinal thinning to reliably detect MCI remains controversial, with some studies reporting pRNFL and GCIPL thinning (33–35), whereas others do not (36– 38). Our study found that participants with MCI had, on average, lower values for pRNFL and GCIPL; these TABLE 4. Vision-specific quality of life scores of MCI participants and controls Group (n) Weighted NEI VFQ-25 10-Item Neuro-Ophthalmic Supplement NEI VFQ-25 plus supplement Control (16) MCI (14) Control (16) MCI (14) Control (16) MCI (14) Mean ± SD 94.4 89.2 89.7 91.3 93.2 89.9 ± ± ± ± ± ± 5.2 7.5 10.2 7.2 5.9 6.0 P† 0.04*‡ 0.98 0.14 *P , 0.05. † Wilcoxon rank-sum test unless otherwise indicated. ‡ Independent samples (2-sample) t test. NEI VFQ-25, National Eye Institute Visual Functioning Questionnaire. Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 85 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 5. Relation of visual test performance with NEIVFQ-25 quality of life scores in participants Visual Measure LCLA 1.25% OU Best MULES time (s) Best MULES errors Correlation Coefficient (rs) P† 0.39 20.30 20.50 0.030* 0.109 0.005** *P , 0.05; **P , 0.01. Spearman rank correlation analysis. OU, oculus uterque. † differences were not significant. Our findings may be a result of the small sample size, or retinal thinning may be less apparent in early disease. Inflammatory changes in the form of retinal ganglion cell swelling and Mueller glial cell hypertrophy are proposed to mask any underlying retinal thinning and contribute to nonsignificant differences or even retinal thickening in early AD compared with controls (39). Interestingly, despite no significant differences in retinal thickness values, MCI participants demonstrated significant reductions at the lowest level of contrast (lightest gray letters on retro-illuminated background) for LCLA. This may suggest that subtle changes in visual function may occur earlier than retinal changes captured by OCT. Alternatively, the OCT measures may have a smaller effect size than other continuous variables (such as LCLA and MULES time scores) in the data set, and thus, a small cohort would impact power even more for this measure. Because this is a preliminary study, there are some important limitations to note. First, we had a relatively small sample size in each group, which decreased our study power to detect differences between groups. Second, most of our MCI participants did not have confirmed amyloid or tau biomarkers, which limits diagnostic certainty. Our MCI participants were also recruited from both a clinic and research cohort and, therefore, were more likely to have potential comorbidities than controls (controls were recruited from a research cohort that underwent strict exclusion for comorbidities). Furthermore, we did not exclude participants with common ophthalmic diseases to preserve a more generalizable cohort of patients with MCI and controls. We did account for comorbid ocular pathology in our analyses; however, severity and type of ocular pathology may vary between groups and affect visual test performance. Finally, mood, which was not accounted for in study analyses, may possibly affect performance on MULES testing. Depressive symptoms are frequently observed in AD dementia and in MCI (40–42), and can impair concentration, attention, memory, and processing speed (43); all of these are needed to accurately perform RAN. CONCLUSIONS The MULES and LCLA testing at 1.25% contrast emerged in this preliminary study as visual measures that distinguish 86 individuals with MCI due to AD vs those with normal cognition; the MRI analyses suggest that these visual measures may be at least complementary to other AD or MCI diagnostic biomarkers. Both MULES and LCLA tests are brief and easily administered, and their scores are associated with vision-specific QOL. Future studies should incorporate a larger cohort, evaluate how these measures compare with other established biomarkers, and investigate the associations of visual pathway structure and function with performance on cognitive testing and disease progression. Investigation in preclinical disease, such as subjective cognitive decline, may elucidate the capacity for vision-based measures to detect even earlier disease processes for patients with AD. REFERENCES 1. El Kadmiri N, Said N, Slassi I, El Moutawakil B, Nadifi S. Biomarkers for Alzheimer disease: classical and novel candidates’ Review. Neuroscience. 2018;370:181–190. 2. Katz B, Rimmer S. Ophthalmologic manifestations of Alzheimer’s disease. Surv Ophthalmol. 1989;34:31–43. 3. McKee AC, Au R, Cabral HJ, Kowall NW, Seshadri S, Kubilus CA, Drake J, Wolf PA. Visual association pathology in preclinical Alzheimer disease. J Neuropathol Exp Neurol. 2006;65:621– 630. 4. Koronyo-Hamaoui M, Koronyo Y, Ljubimov AV, Miller CA, Ko MK, Black KL, Schwartz M, Farkas DL. Identification of amyloid plaques in retinas from Alzheimer’s patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. Neuroimage. 2011;54(Suppl 1):S204–S217. 5. Javaid FZ, Brenton J, Guo L, Cordeiro MF. Visual and ocular manifestations of Alzheimer’s disease and their use as biomarkers for diagnosis and progression. Front Neurol. 2016;7:55. 6. Polo V, Rodrigo MJ, Garcia-Martin E, Otin S, Larrosa JM, Fuertes MI, Bambo MP, Pablo LE, Satue M. Visual dysfunction and its correlation with retinal changes in patients with Alzheimer’s disease. Eye (Lond). 2017;31:1034–1041. 7. Yang Q, Wang T, Su N, Xiao S, Kapoula Z. Specific saccade deficits in patients with Alzheimer’s disease at mild to moderate stage and in patients with amnestic mild cognitive impairment. Age (Dordr). 2013;35:1287–1298. 8. Chan VTT, Sun Z, Tang S, Chen LJ, Wong A, Tham CC, Wong TY, Chen C, Ikram MK, Whitson HE, Lad EM, Mok VCT, Cheung CY. Spectral-domain OCT measurements in Alzheimer’s disease: a systematic Review and meta-analysis. Ophthalmology. 2019;126:497–510. 9. Coppola G, Di Renzo A, Ziccardi L, Martelli F, Fadda A, Manni G, Barboni P, Pierelli F, Sadun AA, Parisi V. Optical coherence tomography in Alzheimer’s disease: a meta-analysis. PLoS One. 2015;10:e0134750. 10. Seay M, Akhand O, Galetta MS, Cobbs L, Hasanaj L, Amorapanth P, Rizzo JR, Nolan R, Serrano L, Rucker JC, Galetta SL, Balcer LJ. Mobile Universal Lexicon Evaluation System (MULES) in MS: evaluation of a new visual test of rapid picture naming. J Neurol Sci. 2018;394:1–5. 11. Akhand O, Galetta MS, Cobbs L, Hasanaj L, Webb N, Drattell J, Amorapanth P, Rizzo JR, Nolan R, Serrano L, Rucker JC, Cardone D, Jordan BD, Silverio A, Galetta SL, Balcer LJ. The new Mobile Universal Lexicon Evaluation System (MULES): a test of rapid picture naming for concussion sized for the sidelines. J Neurol Sci. 2018;387:199–204. 12. Conway J, Ilardi M, Gonzalez C, Dahan N, Fallon S, Moehringer N, Hasanaj L, Joseph B, Serrano L, Rizzo JR, Rucker JC, Feigin A, Frucht S, Galetta SL, Balcer LJ. Rapid picture naming in Parkinson’s disease using the mobile universal lexicon evaluation system (MULES). J Neurol Sci. 2020;410:116680. Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 13. Galetta KM, Chapman KR, Essis MD, Alosco ML, Gillard D, Steinberg E, Dixon D, Martin B, Chaisson CE, Kowall NW, Tripodis Y, Balcer LJ, Stern RA. Screening utility of the kingdevick test in mild cognitive impairment and Alzheimer disease dementia. Alzheimer Dis Assoc Disord. 2017;31:152–158. 14. Dahan N, Moehringer N, Hasanaj L, Serrano L, Joseph B, Wu S, Nolan-Kenney R, Rizzo JR, Rucker JC, Galetta SL, Balcer LJ. The SUN test of vision: investigation in healthy volunteers and comparison to the mobile universal lexicon evaluation system (MULES). J Neurol Sci. 2020;415:116953. 15. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, Phelps CH. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on AgingAlzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:270– 279. 16. Polo V, Garcia-Martin E, Bambo MP, Pinilla J, Larrosa JM, Satue M, Otin S, Pablo LE. Reliability and validity of Cirrus and Spectralis optical coherence tomography for detecting retinal atrophy in Alzheimer’s disease. Eye (Lond). 2014;28:680– 690. 17. Nolan-Kenney RC, Liu M, Akhand O, Calabresi PA, Paul F, Petzold A, Balk L, Brandt AU, Martínez-Lapiscina EH, Saidha S, Villoslada P, Al-Hassan AA, Behbehani R, Frohman EM, Frohman T, Havla J, Hemmer B, Jiang H, Knier B, Korn T, Leocani L, Papadopoulou A, Pisa M, Zimmermann H, Galetta SL, Balcer LJ. Optimal intereye difference thresholds by optical coherence tomography in multiple sclerosis: an international study. Ann Neurol. 2019;85:618–629. 18. Mangione CM, Lee PP, Gutierrez PR, Spritzer K, Berry S, Hays RD. Development of the 25-item national eye Institute visual function questionnaire. Arch Ophthalmol. 2001;119:1050– 1058. 19. Raphael BA, Galetta KM, Jacobs DA, Markowitz CE, Liu GT, Nano-Schiavi ML, Galetta SL, Maguire MG, Mangione CM, Globe DR, Balcer LJ. Validation and test characteristics of a 10-item neuro-ophthalmic supplement to the NEI-VFQ-25. Am J Ophthalmol. 2006;142:1026–1035. 20. Mangione C. The National Eye Institute 25-item Visual Function Questionnaire (VFQ-25) Version 2000 2020. Available at: https://www.nei.nih.gov/learn-about-eye-health/resourcesfor-health-educators/outreach-materials/visual-functionquestionnaire-25. Accessed August 1, 2020. 21. Cohen J. A power primer. Psychol Bull. 1992;112:155–159. 22. Elliott DB, Hurst MA, Weatherill J. Comparing clinical tests of visual function in cataract with the patient’s perceived visual disability. Eye (Lond). 1990;4:712–717. 23. Marron JA, Bailey IL. Visual factors and orientation-mobility performance. Am J Optom Physiol Opt. 1982;59:413–426. 24. Mowry EM, Loguidice MJ, Daniels AB, Jacobs DA, Markowitz CE, Galetta SL, Nano-Schiavi ML, Cutter GR, Maguire MG, Balcer LJ. Vision related quality of life in multiple sclerosis: correlation with new measures of low and high contrast letter acuity. J Neurol Neurosurg Psychiatry. 2009;80:767–772. 25. Kelley BJ, Petersen RC. Alzheimer’s disease and mild cognitive impairment. Neurol Clin. 2007;25:577–609. 26. Akhand O, Rizzo JR, Rucker JC, Hasanaj L, Galetta SL, Balcer LJ. History and future directions of vision testing in head trauma. J Neuroophthalmol. 2019;39:68–81. 27. Devlin JT, Price CJ. Perirhinal contributions to human visual perception. Curr Biol. 2007;17:1484–1488. 28. Barense MD, Henson RN, Lee AC, Graham KS. Medial temporal lobe activity during complex discrimination of faces, Wu et al: J Neuro-Ophthalmol 2022; 42: 79-87 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. objects, and scenes: effects of viewpoint. Hippocampus. 2010;20:389–401. Braak H, Braak E. Neuropathological stageing of Alzheimerrelated changes. Acta Neuropathol. 1991;82:239–259. Krumm S, Kivisaari SL, Probst A, Monsch AU, Reinhardt J, Ulmer S, Stippich C, Kressig RW, Taylor KI. Cortical thinning of parahippocampal subregions in very early Alzheimer’s disease. Neurobiol Aging. 2016;38:188–196. Sone D, Imabayashi E, Maikusa N, Okamura N, Furumoto S, Kudo Y, Ogawa M, Takano H, Yokoi Y, Sakata M, Tsukamoto T, Kato K, Matsuda H. Regional tau deposition and subregion atrophy of medial temporal structures in early Alzheimer’s disease: a combined positron emission tomography/magnetic resonance imaging study. Alzheimers Dement (Amst). 2017;9:35–40. Cummine J, Szepesvari E, Chouinard B, Hanif W, Georgiou GK. A functional investigation of RAN letters, digits, and objects: how similar are they? Behav Brain Res. 2014;275:157–165. Ferrari L, Huang SC, Magnani G, Ambrosi A, Comi G, Leocani L. Optical coherence tomography reveals retinal neuroaxonal thinning in frontotemporal dementia as in Alzheimer’s disease. J Alzheimers Dis. 2017;56:1101–1107. Oktem EO, Derle E, Kibaroglu S, Oktem C, Akkoyun I, Can U. The relationship between the degree of cognitive impairment and retinal nerve fiber layer thickness. Neurol Sci. 2015;36:1141–1146. Wu Y, Wang XN, Wang N, Han Y, Ma D, Lu Y. Regularity changes of the retinal nerve fiber layer and macular ganglion cell complex in patients with the amnestic mild cognitive impairment. Int J Neurosci. 2018;128:849–853. Kim JI, Kang BH. Decreased retinal thickness in patients with Alzheimer’s disease is correlated with disease severity. PLoS One. 2019;14:e0224180. Almeida ALM, Pires LA, Figueiredo EA, Costa-Cunha LVF, Zacharias LC, Preti RC, Monteiro MLR, Cunha LP. Correlation between cognitive impairment and retinal neural loss assessed by swept-source optical coherence tomography in patients with mild cognitive impairment. Alzheimers Dement (Amst). 2019;11:659–669. Pillai JA, Bermel R, Bonner-Jackson A, Rae-Grant A, Fernandez H, Bena J, Jones SE, Ehlers JP, Leverenz JB. Retinal nerve fiber layer thinning in Alzheimer’s disease: a case-control study in comparison to normal aging, Parkinson’s disease, and nonAlzheimer’s dementia. Am J Alzheimers Dis Other Demen. 2016;31:430–436. Ascaso FJ, Cruz N, Modrego PJ, Lopez-Anton R, Santabárbara J, Pascual LF, Lobo A, Cristóbal JA. Retinal alterations in mild cognitive impairment and Alzheimer’s disease: an optical coherence tomography study. J Neurol. 2014;261:1522– 1530. Lyketsos CG, Carrillo MC, Ryan JM, Khachaturian AS, Trzepacz P, Amatniek J, Cedarbaum J, Brashear R, Miller DS. Neuropsychiatric symptoms in Alzheimer’s disease. Alzheimers Dement. 2011;7:532–539. Lyketsos CG, Lopez O, Jones B, Fitzpatrick AL, Breitner J, DeKosky S. Prevalence of neuropsychiatric symptoms in dementia and mild cognitive impairment: results from the cardiovascular health study. JAMA. 2002;288:1475–1483. Ismail Z, Elbayoumi H, Fischer CE, Hogan DB, Millikin CP, Schweizer T, Mortby ME, Smith EE, Patten SB, Fiest KM. Prevalence of depression in patients with mild cognitive impairment: a systematic Review and meta-analysis. JAMA Psychiatry. 2017;74:58–67. Culpepper L, Lam RW, McIntyre RS. Cognitive impairment in patients with depression: Awareness, assessment, and management. J Clin Psychiatry. 2017;78:1383–1394. 87 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.