Journal of Endocrinological Investigation (2023) 46:1875–1880 https://doi.org/10.1007/s40618-023-02046-y ORIGINAL ARTICLE Retinal sensitivity and gaze fixation evaluated by microperimetry in subjects with type 2 diabetes: two independent parameters that explore different neuronal circuits A. M. Ortiz‑Zuñiga1,2 · A. Rojano Toimil1 · K. Rahnama3 · E. Lainez3 · N. Raguer3 · O. Simó‑Servat1,2,4 · C. Hernández1,2,4 · R. Simó1,2,4 · A. Ciudin1,2,4 Received: 25 October 2022 / Accepted: 19 February 2023 / Published online: 4 March 2023 © The Author(s) 2023 Abstract Background and aims Retinal sensitivity (RS) and gaze fixation (GF) assessed by retinal microperimetry are useful and complementary tools for identifying mild cognitive impairment (MCI) in patients with type 2 diabetes (T2D). The hypothesis is that RS and GF examine different neural circuits: RS depends only on the visual pathway while GF reflects white matter complex connectivity networks. The aim of the study is to shed light to this issue by examining the relationship of these two parameters with visual evoked potentials (VEP), the current gold standard to examine the visual pathway. Materials and methods Consecutive T2D patients > 65 years were recruited from the outpatient clinic. Retinal microperim- etry (MAIA 3rd generation) and visual evoked potentials (VEP) (Nicolet Viking ED). RS (dB), GF (BCEA63%, BCEA95%) (MAIA) and VEP (Latency P100ms, Amplitude75–100 uV) were analyzed. Results Thirty three patients (45% women, 72.1 ± 4.6 years) were included. VEP parameters significantly correlated with RS but not with GF. Conclusions These results confirm that RS but not GF depends on the visual pathway, reinforcing the concept that they are complementary diagnostic tools. Used together can further increase the value of microperimetry as screening test for identifying T2D population with cognitive impairment. Keywords Retinal microperimetry · Retinal sensitivity · Gaze fixation · Type 2 diabetes · Visual evoked potentials Introduction Over 50 million people worldwide are affected by dementia, Alzheimer’s disease (AD) being the most common type [1]. * R. Simó Mild cognitive impairment (MCI), is an intermediate stage rafael.simo@vhir.org between normal cognitive function and dementia, which is * A. Ciudin diagnosed by standard tests and has no significant impact on andreea.ciudin@vallhebron.cat activities of daily living. MCI affects about 6% of the general 1 Department of Endocrinology and Nutrition, Hospital population [2], and 20% of the patients > 65 years old [3]. Universitari Vall Hebron, Pg Vall Hebron 119‑129, The annual conversion rate from MCI to dementia ranges 08035 Barcelona, Spain between 10 and 30% in the general population [4]. Patients 2 Institut de Recerca Vall d’Hebron, Universitat Autònoma with type 2 diabetes (T2D) have about 2.5 higher risk of de Barcelona (VHIR-UAB), Pg Vall Hebron 119‑129, developing dementia in comparison with age-matched non- 08035 Barcelona, Spain 3 diabetic subjects [5]. This increased risk is maintained even Department of Clinical Neurophysiology, Hospital after adjusting for vascular risk factors [6, 7]. Moreover, Universitari Vall Hebron, Pg Vall Hebron 119‑129, 08035 Barcelona, Spain the number of cases of dementia associated with T2D is 4 expected to increase because of the increase in prevalence of CIBER de Diabetes y Enfermedades Metabólicas Asociadas, Instituto de Salud Carlos III, C/Monforte de Lemos diabetic individuals due to a worldwide ageing population. 3‑5.Pabellón 11, 28029 Madrid, Spain 13 Vol.:(0123456789) 1876 Journal of Endocrinological Investigation (2023) 46:1875–1880 The American Diabetes Association (ADA) recommends included. The study was conducted according to the Declara- screening for early detection of MCI in patients with T2D tion of Helsinki and was approved by the local Ethics Com- older than 65 years [8]. Currently, the diagnosis of cognitive mittee (PR(AG)28/2017). Informed consent was obtained impairment is hampered by an extensive series of complex from all subjects involved in the study. tests that are difficult to incorporate into daily clinical practice The inclusion criteria were: (1) age > 65 years; (2) T2D [9]. with a duration > 5 years; (3) no apparent or only mild non- Retinal microperimetry, is a simple, rapid and non-invasive proliferative diabetic retinopathy (DR), according to the test that measures retinal sensitivity (RS) in terms of the mini- Early Treatment Diabetic Retinopathy Study (ETDRS) [15] mum light intensity that patients can perceive when spots of (4) patients with normal visual acuity (0.8–1). The exclusion light stimulate specific areas of the retina, and also evaluates criteria were: (1) patients with neurodegenerative diseases gaze fixation (GF) stability [10, 11]. Our group showed for of (e.g., Alzheimer’s disease, Parkinson disease, glaucoma, the first time that the RS significantly correlated with brain multiple sclerosis); (2) patients with HbA1c > 10% (high imaging (MRI and PET) and identified patients with MCI and blood glucose levels could affect retinal function [16] or dementia, as confirmed by complete and validated neuropsy- hyperglycemic or hypoglycemic decompensations in the chological battery examination [10]. last 6 months. (3) Unable or unwilling to perform VEP and Furthermore, by adding the GF parameters to the RS the retinal microperimetry. (4) Patients with any other ophthal- probability to identify cognitive impairment significantly mological disease that could affect the macular integrity. increased in an independent manner, with a sensitivity of The presence of diabetic retinopathy was independently 72.7% and sensibility of 87.9% [12], converting the retinal evaluated by an ophthalmologist by means of fundoscopy microperimetry into a potential useful tool for the screen- and SD-OCT. Furthermore, during the same visit any fovea ing of MCI in patients with T2D > 65 years. Additionally, or macular alterations were excluded. The visual acuity was we have recently evaluated the usefulness of retinal micro- evaluated by means of Snellen chart and was quantified as perimetry as a monitoring tool for the cognitive function in 1.00 (0.0 logMAR) as a criterion for a normal VA in adults. patients > 65 years with T2D [13]. The annual worsening in The VA measurement was always performed without opti- cognitive function went in parallel with worsening of GF, sug- cal compensation of the patients, or with their usual optical gesting this parameter can be a reliable and subtler tool for the compensation, in the case of those patients who use glasses monitoring of the cognitive function in patients with T2D. Our or contact lenses. Subjective refraction was not performed data suggest that since RS is a reliable screening tool for diag- to determine the refractive status of the patient, as it was not nosis, the evaluation of GF could represent a better biomarker part of the study protocol. for annual follow-up. These findings could be explained by the RS was evaluated by fundus-driven microperimetry hypothesis that RS and GF are examining different neural cir- (third-generation Macular Integrity Assessment) after pupil- cuits. Whereas RS depends on the visual pathway, GF reflects lary dilation to a minimum of 4 mm and in a dark envi- the complex white matter connectivity, making them comple- ronment. The standard Macular Integrity Assessment test mentary tests in identifying patients with MCI. Visual evoked covers a 10-diameter area with 37 measurement points; a Potentials (VEP) represents an electrophysiological test used red 1-radius circle was used as the fixation target. A four- to quantify the functional integrity of the visual pathway and level fixed strategy was used: Goldmann III size stimulus, occipital cortex activity, even more sensitive and economical background luminance of 4 asb and maximum luminance of compared to MRI in detecting lesions of the visual pathway 1000 asb, with a 25-dB dynamic range. We chose this strat- [14], and it considered the current gold-standard. Our hypoth- egy to be in line with our previous studies [10, 12]. Patients esis is that RS will correlate with VEP parameters while GF were instructed to maintain the gaze fixed on the red radius will not. To the best of our knowledge this hypothesis was circle as fixation target and press the control button when not explored so far. On these bases we performed the present on the 37 light stimuli appeared. Notably, the microperim- pilot study aimed at exploring direct correlations between the eter automatically compensates for eye movements during VEP parameters and retinal microperimetry (RS and GF) in examination via a software module that tracks them. The patients with T2D > 65 years with no overt or only mild dia- characteristics of fixation (location and stability) were quan- betic retinopathy. tified and categorized according to the P1 and P2 param- eters, the preferred retinal locus (PRL) used to calculate the fixation stability, as follows: “stable” (P1 and P2 0.75%), Materials and methods “relatively unstable” (P1,75% and P2 0.75%), or unstable (both P1 and P2,75%). The device automatically calculates Consecutive patients with T2D older than 65 years attended the reliability index, which assesses the accuracy of the test. at the outpatient clinic of our hospital between June and Furthermore, according to our previous results [10, 13] we December 2019 accomplishing the inclusion criteria were evaluated as part of GF: the bivariate contour ellipse area, 13 Journal of Endocrinological Investigation (2023) 46:1875–1880 1877 as the preferred retinal loci to calculate fixation stability Statistical analysis (BCEA): BCEA63% (1 standard deviation) and BCEA95% (2 standard deviations) areas. The BCEA is measured as a To assess differences between the groups, a x2 test was used bivariate contour ellipse, expressed in “º2”. for qualitative variables and ANOVA followed by a Least It should be noted that our study was aimed to evalu- Significant Difference (LSD) post hoc test were used for ate correlations between two functional independent tests quantitative variables. To evaluate the correlation between (MAIA and VEP) in the same patient and not to test for VEP and RS and GF variables, a Spearman correlation test ophthalmologic disease. We are exploring the potential of and regression analyses adjusted by age were performed. MAIA as a reliable tool for cognitive impairment screening All p values were based on a two-tail statistical significance and monitoring. For this purpose, in the previous studies we test. Significance was accepted at p =  < 0.05. The vari- only analyzed the data from the first eye that was evaluated, ables with normal distribution were: age, Body mass index, to avoid “learning-effect” in the second eye. For the same T2D duration, HbA1C, Retinal microperimetry parameters reasons, in the present study we applied the same procedure, (Sensitivity, Fixation stability P1 (%), Fixation stability including in the analysis data from the right eye in all cased. P2 (%), BCEA63% (º2), BCEA95% (º2)) and Evoked oph- Visual evoked potentials (VEP) were recorded in all thalmic potentials parameters such us Duration (minutes), patients using a Viking Select (Viasys Healthcare, Madison, Latency P100 (ms) 12 × 16; 24 × 34 and Amplitude 75–100 WI, USA) following the guidelines of the American Clinical (uV) 12 × 16; 24 × 34. All the correlations were adjusted by Neurophysiology Society [17] and ISCEV recommendations gender, age, duration of T2D and smoking status. Statisti- [18]. The visual stimulus was applied using Nicolet 2015 cal analysis was performed with the STATA 15 statistical Visual Stimulator (CareFusion NeuroCare, Middleton, WI, package. USA) with a monitor size of 17 inches in all recordings. Each eye was stimulated separately using a screen full pat- tern reversal chromatic checks of mid-size (12X16; visual Results angle = 1° 0′) and small checks (24 × 32; visual angle = 0° 30′). Full-field stimulation was performed monocularly, uti- A total of 33 patients with T2D (45% women, mean age lizing a high-contrast (> 50%) black-and-white checkerboard 72.1 ± 4.6 years) were recruited fulfilling inclusion/exclu- pattern, at a rate of 1.3 cycles per second. The subject was sion criteria. The characteristics of the patients are shown placed 107 cm to the stimulus screen. Visual fixation was in Table 1. at the center of the stimulus screen. Recordings were made VEP parameters significantly correlated with RS evalu- using surface electrodes fastened securely over Oz, with Fz ated by microperimetry as reflected by Table 2. By contrast, as the reference, according to the 10–20 international system we did not find any significant correlations between the GF of EEG electrode placement. A ground electrode was placed parameters P1 and BCEA95% and VEP, respectively. on the forehead. The impedance used in all cases was less than 4000 Ω. The low filter was set at 1 Hz and the high filter at 100 Hz. Recordings were made during the first 500 ms Discussion after the stimulus. At least 100 steps were averaged for each tracing, and 2 or 3 replications were run and superimposed In the present study we provide first evidence that RS to ensure reproducibility. Latency of the P100 component but not GF assessed by microperimetry is correlated and amplitude N75-P100 were measured for both each eye with visual pathway explored by VEP in patients with and each check size. The parameters evaluated for VEP type 2 diabetes, older than 65 years without macular or for each check size wer: (1) the presence or absence and foveal affectation. As far as we know one study evalu- latency of wave P100; (2) the N75-P100 amplitude; and (3) ated the relationship between the VEP and microperim- symmetry between both eyes. P100 wave is the most robust etry in patients with macular disease and found corre- peak with comparatively minimal interindividual variability, lation between RS and VEP in 48 young (45 years old) nominal within-subject inter-eye difference, and negligible patients with altered fixation pattern [19]. The present variation with high repeatability [16]. P100 values from the study provides first data regarding this relationship in right eye were used for correlations. subjects without macular disease. Our findings are of interest in the context of proposing the microperimetry a Sample size calculation useful tool for screening and monitoring of the cognitive function apart from its current ophthalmologic applica- This was a pilot study. Since there is no previous data aimed tion [10–12]. In the last years the research from our group to explore similar hypothesis, we could not perform sample focused on demonstrating that retinal microperimetry is size calculation. useful for screening for cognitive impairment in patients 13 1878 Journal of Endocrinological Investigation (2023) 46:1875–1880 Table 1  Baseline characteristics of the patients included in the study Additionally, in a recent study [13], we showed that GF (n = 33) evaluated by microperimetry was useful for the annual Clinical characteristics monitoring of the cognitive function, while RS remained unchanged in time, supporting the same hypothesis. The Age (years) mean ± SD 72.1 ± 4.6 results from this pilot study are the first technical piece of Gender (women) % 45% evidence towards proving that GF and RS are independent BMI (kg/m2) mean ± SD 29.74 ± 0.9 parameters, since RS correlated with the VEP and GF was Smoker (%) 59.9% not. Both parameters, RS and GF can be considered com- Hypertension (%) 84.8% plementary measurements: RS will be more influenced by Dyslipidemia (%) 87.9% neurodegeneration while GF by white matter disruption, Sleep apnea syndrome (%) 16.6% as we explain further. In the case of RS, the anatomical Ischemic heart disease (%) 18.1% region involved in the processing of the information that Peripheral arteriopathy (%) 6.0% goes through the optic pathway from the retinal ganglion T2D duration (years, mean ± SD) 15.55 ± 7.4 cells, is the lateral geniculate body of the thalamus, which HbA1C (%) (DCCT mean ± SD) 7.38 ± 0.8 relays information to the primary visual cortex [16]. To Diabetic retinopathy (%) 36.3% support out findings, Arruda et al. [20] has recently sug- Diabetic nephropathy (%) 27.2% gested the potential role of VEP in detecting amnestic MCI Diabetic polyneuropathy (%) 18.1% that precedes Alzheimer´s disease, with primary dysfunc- Visual acuity (1.00) 0.85 tion of the episodic memory, which is related to neuro- Retinal microperimetry (right eye) degeneration, allowing a connection between the optical Sensitivity (mean ± SD) 20.75 ± 0.9 pathway and cortical cognitive function. Additionally, Duration (minutes) (mean ± SD) 2.55 ± 0.4 RS seems to reflect early alterations in the primary visual Reliability index % (mean ± SD) 98.1 ± 1.1 cortex. McKee et al. [21] demonstrated the presence of Fixation stability P1 (%) (mean ± SD) 70.81 ± 13.3 amyloid plaques and neurofibrillary tangles in areas were Fixation stability P2 (%) (mean ± SD) 84.07 ± 4.7 VEP are generated, before than other areas in the evolution BCEA63%(º2) (mean ± SD) 3.67 ± 1.6 of Alzheimer’s disease. Therefore, it is reasonable that we BCEA95%(º2) (mean ± SD) 32.62 ± 9.3 have found a correlation between RS explored by retinal Evoked ophthalmic potentials (right eye) microperimetry and VEP. We could hypothesize that RS Duration (minutes) 15.16 ± 1.41 and VEP explore the same neural circuits, nevertheless, Latency P100 (ms)12 × 16 (mean ± SD) 106.43 ± 11.4 future studies could shed light on this interesting findings. Latency P100 (ms)24 × 34 (mean ± SD) 111.97 ± 10.8 In exchange, GF seems to depend on the complex white Amplitude 75–100 (uV) 12 × 16 (mean ± SD) 6.17 ± 2.8 matter network, the superior colliculus and the parietal Amplitude 75–100 (uV) 24 × 34 (mean ± SD) 5.94 ± 2.81 and frontal cortex [22]. Additionally, in a previous study BMI body mass index, T2D type 2 diabetes, BCEA63 Bivariate con- published by Ortiz et al. [13], GF but not RS was cor- tour ellipse area 63% (º2), BCEA95 Bivariate contour ellipse area 95% related with attention and delayed recall, which depend (º2) mainly on the default brain network and the white matter connections between the dorsolateral and prefrontal cortex with T2D > 65 years [10, 13]. Furthermore, both param- and inferior and superior parietal lobules [23, 24]. Fur- eters, RS and GF when used together increase the ability thermore, we also found that GF was impaired in young to correctly identify patients with T2D > 65 years and mild patients with obesity and significantly correlated with cog- cognitive impairment or dementia, suggesting that they nitive function, in particular attention and delayed recall, are independent parameters that explore different circuits. all of them related to white matter network and not visual Table 2  Correlations between Correlations between microperimetry and VEP Latency 100 ms Amplitude microperimetry and VEP (75–100 uV) parameters Retinal sensitivity (dB) R = 0.716* R = 0.531* Fixation stability P1 (%) R = 0.175, R = 0.109 Fixation stability P2 (%) R = 0.151 R = 0.119 Bivariate contour ellipse area 95% (º2) R = 0.156 R = 0.238 Bivariate contour ellipse area 63% (º2) R = 0.293 R = 0.127 *p < 0.05 13 Journal of Endocrinological Investigation (2023) 46:1875–1880 1879 pathway [12]. Therefore, the present observation that fixa- Funding Open Access Funding provided by Universitat Autonoma de tion parameters evaluated by retinal microperimetry did Barcelona. “This study was supported by grants from the Instituto de Salud Carlos III (Fondo de Investigación Sanitaria, PI20/01703). The not correlate with VEP in our study is not surprising. funders had no role in study design, data collection and analysis, deci- As explained in the introduction, the American Dia- sion to publish, or preparation of the manuscript.” betes Association recommends annual evaluation of the cognitive function in patients with T2D older than Data availability Data is available on request to the corresponding authors. 65 years [8]. Furthermore, it is unfeasible to implement a complete battery of neuropsychological tests for all T2D Declarations patients > 65 years because they are time consuming and also depend on the patient´s mood, therefore we urgently Conflict of interest The authors declare that they have no conflict of need simple and reliable tools. Retinal microperimetry, interest. is a simple, objective and rapid test that can be largely Institutional review board statement The study was conducted accord- proposed for implementation for the monitoring of cog- ing to the guidelines of the Declaration of Helsinki, and approved by nitive performance, regardless of psychological status or the Institutional Review Board (or Ethics Committee) of Vall Hebron educational level, and can provide two independent and Research Institute (protocol code PR(AG)28/2017. complementary parameters: RS and GF. In this regard, our Research involving human participants and/or animals The study group is coordinating an ongoing European project (REC- involved human participants. OGNISED study, NCT04281186), aimed at validating the usefulness of retinal microperimetry as reliable screening Informed consent Informed consent was obtained from all subjects involved in the study. and monitoring tools for cognitive impairment in patients with T2D > 65 years, as well as at exploring its predictive Open Access This article is licensed under a Creative Commons Attri- value in the progression of the cognitive decline. bution 4.0 International License, which permits use, sharing, adapta- Nevertheless, our study has three main limitations. tion, distribution and reproduction in any medium or format, as long First, the small sample size can be contemplated as a lim- as you give appropriate credit to the original author(s) and the source, iting factor. However, the pilot nature and the design of provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are the study (intra-individual comparisons) makes possible included in the article's Creative Commons licence, unless indicated to draw valid results. Second, we have not compared our otherwise in a credit line to the material. If material is not included in results with any normative data base. However, it should the article's Creative Commons licence and your intended use is not be noted that normal values of microperimetry and VEP permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a in subjects > 65 years has not been reported. In addition, copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. the aim of the study was not to perform these comparisons but to examine correlations between VEP and either RS and GF to further understand the physiological meaning of these examinations. Third, although the 4 level-fixed References strategy used in our study for retinal microperimetry, has 1. Prince M, Bryce R, Albanese E, Wimo A, Ribeiro W, Ferri CP large difference of dB between one step and another, we (2013) The global prevalence of dementia: a systematic review still found significant correlation between RS and VEP. and metaanalysis. Alzheimers Dement J Alzheimers Assoc In conclusion, our results confirm that RS but not 9(1):63-75.e2 2. Sachdev PS, Lipnicki DM, Kochan NA, Crawford JD, Thala- GF depends on the visual pathway, reinforcing the con- muthu A, Andrews G et al (2015) The prevalence of mild cogni- cept that they are complementary diagnostic tools. Used tive impairment in diverse geographical and ethnocultural regions: together can further increase the value of microperimetry the COSMIC collaboration. PLoS ONE 10(11):e0142388 as screening test for identifying T2D population with cog- 3. Lopez OL, Kuller LH, Becker JT, Dulberg C, Sweet RA, Gach HM et al (2007) Incidence of dementia in mild cognitive impair- nitive impairment. However, further studies are needed in ment in the cardiovascular health study cognition study. Arch order to validate our results in larger cohorts of patients Neurol 64(3):416–420 and to deepen the underlying circuits and mechanisms 4. Alegret M, Cuberas-Borrós G, Vinyes-Junqué G, Espinosa A, involved. Valero S, Hernández I et al (2012) A two-year follow-up of cogni- tive deficits and brain perfusion in mild cognitive impairment and mild Alzheimer’s disease. J Alzheimers Dis JAD 30(1):109–120 5. Ott A, Stolk RP, van Harskamp F, Pols HA, Hofman A, Breteler Author contributions “Conceptualization, AC and RS; methodology, MM (1999) Diabetes mellitus and the risk of dementia: the Rot- AC, AO, AR, RS, NR; software, KR, EL; validation, AC, RS, CH; for- terdam study. Neurology 53(9):1937–1942 mal analysis, AO, AR; investigation, AO, AR, KR, EL, OS; resources, 6. Ciudin A, Espinosa A, Simó-Servat O, Ruiz A, Alegret M, AC, NR.; data curation, AO, AR; writing—original draft preparation, Hernández C et al (2017) Type 2 diabetes is an independent risk AO, AR; writing—review and editing, AC, RS; visualization, AC, RS factor for dementia conversion in patients with mild cognitive CH; supervision, AC; project administration, AC. impairment. J Diabetes Complic 31(8):1272–1274 13 1880 Journal of Endocrinological Investigation (2023) 46:1875–1880 7. Wang KC, Woung LC, Tsai MT, Liu CC, Su YH, Li CY (2012) 17. American Clinical Neurophysiology Society (2006) Guideline 9B: Risk of Alzheimer’s disease in relation to diabetes: a population- Guidelines on visual evoked potentials. J Clin Neurophysiol Off based cohort study. Neuroepidemiology 38(4):237–244 Publ Am Electroencephalogr Soc 23(2):138–156 8. American Diabetes Association (2020) Older adults: stand- 18. Robson AG, Nilsson J, Li S, Jalali S, Fulton AB, Tormene AP et al ards of medical care in diabetes-2020. Diabetes Care 43(Suppl (2018) ISCEV guide to visual electrodiagnostic procedures. Doc 1):152–162 Ophthalmol Adv Ophthalmol 136(1):1–26 9. Langa KM, Levine DA (2014) The diagnosis and manage- 19. Jiang L, Zhang H, Xie J, Jiao X, Zhou H, Ji H et al (2011) ment of mild cognitive impairment: a clinical review. JAMA Application of multifocal visual evoked potentials in the assess- 312(23):2551–2561 ment of visual dysfunction in macular diseases. Eye Lond Engl 10. Ciudin A, Simó-Servat O, Hernández C, Arcos G, Diego S, 25(10):1302–1309 Sanabria Á et al (2017) Retinal microperimetry: a new tool for 20. Arruda JE, McInnis MC, Steele J (2020) The flash visual evoked identifying patients with type 2 diabetes at risk for developing potential-P2 and the detection of amnestic mild cognitive impair- Alzheimer disease. Diabetes 66(12):3098–3104 ment: a review of empirical literature. Int J Psychophysiol Off J 11. Wu Z, Ayton LN, Guymer RH, Luu CD (2014) Comparison Int Organ Psychophysiol 155:162–167 between multifocal electroretinography and microperimetry in 21. McKee AC, Au R, Cabral HJ, Kowall NW, Seshadri S, Kubilus age-related macular degeneration. Invest Ophthalmol Vis Sci CA et al (2006) Visual association pathology in preclinical Alz- 55(10):6431–6439 heimer disease. J Neuropathol Exp Neurol 65(6):621–630 12. Simó-Servat O, Ciudin A, Ortiz-Zúñiga ÁM, Hernández C, Simó 22. Klemp K, Larsen M, Sander B, Vaag A, Brockhoff PB, Lund- R (2019) Usefulness of Eye fixation assessment for identifying Andersen H (2004) Effect of short-term hyperglycemia on multi- type 2 diabetic subjects at risk of dementia. J Clin Med 8(1):59 focal electroretinogram in diabetic patients without retinopathy. 13. Ortiz-Zúñiga ÁM, Simó-Servat O, Rojano-Toimil A, Vázquez-de Invest Ophthalmol Vis Sci 45(10):3812–3819 Sebastian J, Castellano-Tejedor C, Hernández C et al (2021) The 23. Bergeron A, Guitton D (2000) Fixation neurons in the superior GF assessed by microperimetry: a useful tool for the monitoring colliculus encode distance between current and desired gaze posi- of the cognitive function in patients with type 2 diabetes. J Pers tions. Nat Neurosci 3(9):932–939 Med 11(8):698 24. Bartolomeo P, Thiebaut de Schotten M, Chica AB (2012) Brain 14. Kothari R, Bokariya P, Singh S, Singh R (2016) A comprehensive networks of visuospatial attention and their disruption in visual review on methodologies employed for visual evoked potentials. neglect. Front Hum Neurosci 6:110 Scientifica 2016:9852194 15. Fundus photographic risk factors for progression of diabetic Publisher's Note Springer Nature remains neutral with regard to retinopathy. ETDRS report number 12. Early Treatment Diabetic jurisdictional claims in published maps and institutional affiliations. Retinopathy Study Research Group. Ophthalmology. 1991. 98(5 Suppl):823–33 16. Parvizi J, Van Hoesen GW, Damasio A (2001) The selective vul- nerability of brainstem nuclei to Alzheimer’s disease. Ann Neurol 49(1):53–66 13