Prospective Study Open Access
Copyright ©The Author(s) 2024. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Diabetes. Apr 15, 2024; 15(4): 697-711
Published online Apr 15, 2024. doi: 10.4239/wjd.v15.i4.697
Association of age at diagnosis of diabetes with subsequent risk of age-related ocular diseases and vision acuity
Si-Ting Ye, The Second Clinical College, Guangzhou University of Chinese Medicine, Guangzhou 510120, Guangdong Province, China
Si-Ting Ye, Department of Ultrasound, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510120, Guangdong Province, China
Xian-Wen Shang, Yu Huang, Zhuo-Ting Zhu, Xue-Li Zhang, Shu-Lin Tang, Xiao-Hong Yang, Ming-Guang He, Guangdong Eye Institute, Department of Ophthalmology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, Guangdong Province, China
Xian-Wen Shang, Yu Huang, Zhuo-Ting Zhu, Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, Guangdong Province, China
Susan Zhu, Austin Hospital, University of Melbourne, Melbourne 3084, Victoria, Australia
Wei Wang, Ming-Guang He, State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou 510060, Guangdong Province, China
Zong-Yuan Ge, Monash e-Research Center, Faculty of Engineering, Airdoc Research, Nvidia AI Technology Research Center, Monash University, Melbourne 3080, Victoria, Australia
Ming-Guang He, Department of Ophthalmic Epidemiology, Centre for Eye Research Australia, Melbourne 3002, Victoria, Australia
ORCID number: Si-Ting Ye (0009-0004-2882-5298); Ming-Guang He (0000-0002-6912-2810).
Co-first authors: Si-Ting Ye and Xian-Wen Shang.
Co-corresponding authors: Xiao-Hong Yang and Ming-Guang He.
Author contributions: Shang XW, Yang XH, and He MG conceived the study; Shang XW did the literature search and analysed the data; Ye ST, Shang XW, Huang Y, Zhu S, Zhu ZT, and He MG contributed to key data interpretation; Ye ST and Shang XW wrote the manuscript; Ye ST, Shang XW, Huang Y, Zhu S, Zhu ZT, Zhang XL, Wang W, Tang SL, Ge ZY, Yang XH, and He MG critically revised the manuscript. Ye ST is a clinician who identified the clinical problem, provided clinical background, and facilitated the discussion. Shang XW is a statistician who conducted the analysis. Additionally, Ye ST and Shang XW collaborated on drafting the manuscript. Therefore, Ye ST and Shang XW made equal contributions to this work and are co-first authors. He MG provided the funding and the data source, while Yang XH established the research team for this project. They jointly supervised this work. Therefore, they are both considered co-corresponding authors.
Supported by National Natural Science Foundation of China, No. 32200545; The GDPH Supporting Fund for Talent Program, No. KJ012020633 and KJ012019530; Science and Technology Research Project of Guangdong Provincial Hospital of Chinese Medicine, No. YN2022GK04.
Institutional review board statement: The UK Biobank Study’s ethical approval had been granted by the National Information Governance Board for Health and Social Care and the NHS North West Multicenter Research Ethics Committee. The data used in this study is available in the UK Biobank database under the application number of 62443.
Informed consent statement: All participants provided informed consent through electronic signature at the baseline assessment.
Conflict-of-interest statement: The authors have no conflicts of interest to declare that are relevant to the content of this article.
Data sharing statement: No additional data are available.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Ming-Guang He, MD, PhD, Professor, Guangdong Eye Institute, Department of Ophthalmology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, No. 106 Zhongshaner Road, Guangzhou 510080, Guangdong Province, China. mingguang_he@yahoo.com
Received: November 19, 2023
Peer-review started: November 19, 2023
First decision: December 8, 2023
Revised: December 19, 2023
Accepted: February 27, 2024
Article in press: February 27, 2024
Published online: April 15, 2024

Abstract
BACKGROUND

The importance of age on the development of ocular conditions has been reported by numerous studies. Diabetes may have different associations with different stages of ocular conditions, and the duration of diabetes may affect the development of diabetic eye disease. While there is a dose-response relationship between the age at diagnosis of diabetes and the risk of cardiovascular disease and mortality, whether the age at diagnosis of diabetes is associated with incident ocular conditions remains to be explored. It is unclear which types of diabetes are more predictive of ocular conditions.

AIM

To examine associations between the age of diabetes diagnosis and the incidence of cataract, glaucoma, age-related macular degeneration (AMD), and vision acuity.

METHODS

Our analysis was using the UK Biobank. The cohort included 8709 diabetic participants and 17418 controls for ocular condition analysis, and 6689 diabetic participants and 13378 controls for vision analysis. Ocular diseases were identified using inpatient records until January 2021. Vision acuity was assessed using a chart.

RESULTS

During a median follow-up of 11.0 years, 3874, 665, and 616 new cases of cataract, glaucoma, and AMD, respectively, were identified. A stronger association between diabetes and incident ocular conditions was observed where diabetes was diagnosed at a younger age. Individuals with type 2 diabetes (T2D) diagnosed at < 45 years [HR (95%CI): 2.71 (1.49-4.93)], 45-49 years [2.57 (1.17-5.65)], 50-54 years [1.85 (1.13-3.04)], or 50-59 years of age [1.53 (1.00-2.34)] had a higher risk of AMD independent of glycated haemoglobin. T2D diagnosed < 45 years [HR (95%CI): 2.18 (1.71-2.79)], 45-49 years [1.54 (1.19-2.01)], 50-54 years [1.60 (1.31-1.96)], or 55-59 years of age [1.21 (1.02-1.43)] was associated with an increased cataract risk. T2D diagnosed < 45 years of age only was associated with an increased risk of glaucoma [HR (95%CI): 1.76 (1.00-3.12)]. HRs (95%CIs) for AMD, cataract, and glaucoma associated with type 1 diabetes (T1D) were 4.12 (1.99-8.53), 2.95 (2.17-4.02), and 2.40 (1.09-5.31), respectively. In multivariable-adjusted analysis, individuals with T2D diagnosed < 45 years of age [β 95%CI: 0.025 (0.009,0.040)] had a larger increase in LogMAR. The β (95%CI) for LogMAR associated with T1D was 0.044 (0.014, 0.073).

CONCLUSION

The younger age at the diagnosis of diabetes is associated with a larger relative risk of incident ocular diseases and greater vision loss.

Key Words: Diabetes, Age at diagnosis, Cataract, Glaucoma, Age-related macular disease, Vision acuity

Core Tip: This is the first prospective cohort study to examine the association of age at the diagnosis of diabetes with main ocular conditions. Our findings suggest the age at the diagnosis of diabetes plays an important role in the association between diabetes and incident cataract, glaucoma, and age-related macular disease as well as vision. A younger age at the diagnosis of diabetes was associated with larger excessive relative risk for ocular conditions and larger vision loss. Type 1 diabetes appears to have potentially more harmful effects.
INTRODUCTION

Although the age-standardised prevalence of avoidable vision impairment did not change, the global number of cases increased substantially due to the increasing aging population[1]. Cataract, glaucoma, and age-related macular degeneration (AMD) are the first, second, and fourth leading global causes of blindness in individuals aged 50 years and older, which accounted for 15.2 million, 3.6 million and 1.8 million cases, respectively[1,2]. Therefore, it is imperative to identify the important determinants for these ocular conditions.

Previous evidence has highlighted the importance of diabetes in the development of ocular conditions[3,4]. Diabetes has been linked to numerous ocular conditions, including cataract[5], glaucoma[6], and AMD[7]. The United Kingdom Million Women Study, involving 1312051 postmenopausal women, demonstrated that diabetes was an important risk factor for cataract surgery[8]. In contrast, evidence suggests diabetes is not among the leading predictors for glaucoma[9,10], and other studies did not find a significant association between diabetes and glaucoma[11]. Previous studies have been inconsistent regarding the association of diabetes with AMD[7]. Several studies have demonstrated a positive relationship between diabetes and AMD[12,13], but more studies did not find a significant association[14-18].

The importance of age on the development of ocular conditions has been reported by numerous studies[5,7,9,10]. Diabetes may have different associations with different stages of ocular conditions[19], and the duration of diabetes may affect the development of diabetic eye disease[3]. While there is a dose-response relationship between the age at diagnosis of diabetes and the risk of cardiovascular disease and mortality[20,21], whether the age at diagnosis of diabetes is associated with incident ocular conditions remains to be explored. It is unclear which types of diabetes are more predictive of ocular conditions.

It is important to identify the life stage at which a diagnosis of diabetes is associated with the highest risk of major ocular conditions for the prevention or screening of these conditions. Using the UK Biobank, we sought to examine the association between age at the diagnosis of diabetes and the incidence of cataract, glaucoma, and AMD.

MATERIALS AND METHODS
Study population

The UK Biobank is a population-based cohort of more than 500000 participants aged 40-73 years at baseline, recruited between 2006 and 2010 from one of the 22 assessment centres across England, Wales, and Scotland[22]. The design and population of the UK Biobank study have been described in detail elsewhere[22]. The UK Biobank Study’s ethical approval had been granted by the National Information Governance Board for Health and Social Care and the NHS North West Multicenter Research Ethics Committee. All participants provided informed consent through electronic signature at the baseline assessment. The data used in this study is available in the UK Biobank database under the application number of 62443.

Population selection for eye disease analysis

Individuals with missing data on self-reported eye health (n = 327891), or those with ocular diseases (n = 26320) at baseline were excluded from the analysis. After the exclusion of individuals with missing values on the age at the diagnosis of diabetes or with other type of diabetes rather than type 1 diabetes (T1D) and type 2 diabetes (T2D, n = 264), 7917 participants with T2D were divided into six groups according to the age at diagnosis: < 45, 45-49, 50-54, 55-59, 60-64, and ≥ 65 years. For each diabetic participant, two controls were randomly selected from those without diabetes at baseline using propensity scores matched by age, gender, ethnicity, education, household income, physical activity, smoking, alcohol consumption, sleep duration, depression, hypertension, heart disease, stroke, body mass index (BMI), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and triglycerides. This analysis was conducted for each diabetes diagnosis age group. The same method was used to randomly select controls for T1D patients (n = 792, Figure 1).

Figure 1
Open in New Tab   Full Size Figure   Download Figure
Figure 1 Flowchart for population selection for analysis of ocular conditions from the UK Biobank. Propensity score matching was to select two controls for each diabetic participant. The analysis was conducted for age groups of diabetes diagnosis separately. The median age at diagnosis type 1 was 17 years. Propensity score accounted for age, gender, ethnicity, education, household income, physical activity, smoking, alcohol consumption, sleep duration, depression, hypertension, heart disease, stroke, body mass index, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglyceride. AMD: Age-related macular degeneration; T1D: Type 1 diabetes; T2D: Type 2 diabetes.
Population selection for vision impairment analysis

Among 117252 individuals who had their vision acuity assessed, 7274 had diabetes at baseline. After excluding individuals with missing values on diabetes diagnosis age or with other type of diabetes rather than T1D/T2D (n = 585), 6192 with T2D were divided into six groups according to the diagnosis age: < 45, 45-49, 50-54, 55-59, 60-64, and ≥ 65 years. The same method was used to select controls for individuals with T1D (Supplementary Figure 1).

Age at diagnosis of diabetes

First, participants were classified as diabetic if they reported that a doctor had ever told them that they had diabetes (Field code: 2443). For those with a self-reported diagnosis of diabetes, they were asked a follow-up question “What was your age when diabetes was first diagnosed?” Participants with a potentially abnormal age at the diagnosis of diabetes were asked to confirm. Algorithms based on self-reported medical history and medication were used to identify T1D and T2D[23]. Furthermore, the codes for international classification diseases (ICD) were used to define T1D/T2D (Supplementary Table 1). The age at the diagnosis of diabetes (years) was then computed by subtracting the birth date from the initial diagnosed date divided by 365.25.

Ascertainment of ocular conditions

Individuals were classified as having AMD (Field code: 1528), cataract (1278), or glaucoma (1277) if they reported a diagnosis of the corresponding conditions. Cases of ocular conditions were also identified using hospital inpatient records based on ICD codes (Supplementary Table 1). Furthermore, we used surgical procedures by OPCS4 to identify cataract events (codes: C71.2 or C75.1)[24]. The onset date of ocular condition was defined as the earliest recorded code date regardless of source. Person-years were calculated from the date of baseline assessment to the date of onset ocular condition, date of death, or the end of follow-up (December 31, 2020 for England and Wales and January 31, 2021 for Scotland), whichever came first.

Vision acuity test

The baseline vision acuity examination was performed among a sub-cohort of the UK Biobank from June 2009 to July 2010. The procedure for the vision acuity test has been described in detail elsewhere[25]. Presenting distance vision acuity was measured at 4 m or at 1 m (if a participant was unable to read) using the logarithm of the minimum angle of resolution (LogMAR) chart on a computer screen. Vision was defined as the presenting vision acuity in the better-seeing eye in the analysis.

Covariates

BMI was calculated as weight in kilograms divided by the square of height in meters. A touchscreen computer was used to collect information, including age, gender, education, income, smoking, alcohol consumption, and sleep duration. Metabolic equivalent-hours/week of physical activity during work and leisure time was estimated using specific questions[26]. Sleep duration was categorized into three groups: < 7, 7-9, and > 9 h[27].

Hypertension, depression, stroke, and heart disease at baseline were defined based on self-reported data. Glycated haemoglobin (HbA1c) was measured using high-performance liquid chromatography on a Bio-Rad Variant II Turbo. Lipids, including total cholesterol, HDL-C, LDL-C, and triglycerides, were measured by direct enzymatic methods (Konelab, Thermo Fisher Scientific, Waltham, Massachusetts).

Statistical analysis

T-test was used to test the difference in continuous variables and Chi-square test in categorical variables between diabetic participants and controls in each diabetes diagnosis age group.

The HR with 95%CIs for incident ocular condition associated with T1D and age at diagnosis of T2D was estimated using Cox proportional hazard regression models. The multivariable analysis included adjustment for matching factors (propensity score) and the full model further incorporated concurrent HbA1c. This analysis was separately conducted for incidence of cataract, cataract surgery, glaucoma, and AMD. The analysis was not performed for types of glaucoma or AMD due to their low incidence.

General linear regression models were used to test the difference in LogMAR between diabetic participants and controls for each diagnosis age group. The multivariable analysis included adjustments for matching factors (propensity score). The association between age at the diagnosis of diabetes and intraocular pressure (IOP) was examined using general linear regression models.

A sensitivity analysis was conducted to examine whether the association between age at the diagnosis of T2D and ocular conditions and vision acuity was independent of duration of diabetes. In this analysis, two controls for each T2D patient were randomly selected using propensity score matching based on the same factors as depicted in Figure 1 and Supplementary Figure 1, without stratification by the age at the diagnosis of diabetes. The age at the diagnosis of T2D, treated as a categorical variable (< 45, 45-49, 50-54, 55-59, 60-64, and ≥ 65 years), was analysed to assess the association between the age at the diagnosis of diabetes and ocular conditions and vision acuity.

Missing values for categorical variables were assigned as a single category. Missing values for continuous covariates were imputed with the mean.

Data analyses were conducted using SAS 9.4 for Windows (SAS Institute Inc.), and all P values were two-sided, with statistical significance set at < 0.05.

RESULTS
Baseline characteristics

For ocular condition analysis, 26127 participants (36.9% females) aged 40-70 (mean ± SD: 59.1 ± 8.2) years old were included in the analysis. Diabetic participants had higher HbA1c, and education levels compared to the controls. No significant difference in other characteristics between the two groups were observed (Table 1). Individuals with T1D had higher HbA1c but did not differ in other characteristics compared to the controls (Supplementary Table 2).

Table 1 Baseline characteristics of participants by diabetes and controls for analysis of ocular conditions.
< 45 yr1
45-49 yr1
50-54 yr1
55-59 yr1
60-64 yr1
≥ 65 yr1
Non-diabetesDiabetesNon-diabetesDiabetesNon-diabetesDiabetesNon-diabetesDiabetesNon-diabetesDiabetesNon-diabetesDiabetes
Age (yr)51.9 ± 8.451.9 ± 7.954.7 ± 8.654.7 ± 5.858.7 ± 7.858.7 ± 4.862.2 ± 5.962.0 ± 3.565.0 ± 3.664.9 ± 2.467.6 ± 2.767.6 ± 1.4
Gender
Female1032 (36.8)506 (36.1)743 (33.4)400 (36.0)1163 (37.0)600 (38.2)1312 (36.0)652 (35.8)1075 (35.8)541 (36.0)402 (39.5)202 (39.7)
Male1774 (63.2)897 (63.9)1481 (66.6)712 (64.0)1979 (63.0)971 (61.8)2330 (64.0)1169 (64.2)1927 (64.2)960 (64.0)616 (60.5)307 (60.3)
Ethnicity
Whites1937 (69.0)968 (69.0)1634 (73.5)812 (73.0)2522 (80.3)1255 (79.9)3225 (88.6)1600 (87.9)2715 (90.4)1352 (90.1)939 (92.2)469 (92.1)
Non-whites785 (28.0)421 (30.0)535 (24.1)296 (26.6)548 (17.4)297 (18.9)367 (10.1)207 (11.4)245 (8.2)141 (9.4)71 (7.0)36 (7.1)
Unknown84 (3.0)14 (1.0)55 (2.5)4 (0.4)72 (2.3)19 (1.2)50 (1.4)14 (0.8)42 (1.4)8 (0.5)8 (0.8)4 (0.8)
Education
0-5 yr863 (30.8)379 (27.0)640 (28.8)266 (23.9)839 (26.7)397 (25.3)907 (24.9)436 (23.9)650 (21.7)318 (21.2)206 (20.2)105 (20.6)
6-12 yr1270 (45.3)687 (49.0)1108 (49.8)597 (53.7)1524 (48.5)776 (49.4)1631 (44.8)853 (46.8)1368 (45.6)681 (45.4)419 (41.2)202 (39.7)
≥ 13 yr567 (20.2)292 (20.8)421 (18.9)225 (20.2)706 (22.5)361 (23.0)1044 (28.7)493 (27.1)927 (30.9)477 (31.8)369 (36.2)192 (37.7)
Missing106 (3.8)45 (3.2)55 (2.5)24 (2.2)73 (2.3)37 (2.4)60 (1.6)39 (2.1)57 (1.9)25 (1.7)24 (2.4)10 (2.0)
Household income (pounds)
< 18000680 (24.2)418 (29.8)575 (25.9)320 (28.8)846 (26.9)470 (29.9)1075 (29.5)550 (30.2)912 (30.4)528 (35.2)372 (36.5)209 (41.1)
18000-30999529 (18.9)285 (20.3)446 (20.1)242 (21.8)696 (22.2)328 (20.9)868 (23.8)445 (24.4)751 (25.0)371 (24.7)267 (26.2)121 (23.8)
31000-51999535 (19.1)244 (17.4)420 (18.9)211 (19.0)576 (18.3)273 (17.4)600 (16.5)297 (16.3)454 (15.1)218 (14.5)110 (10.8)55 (10.8)
52000-100000369 (13.2)152 (10.8)319 (14.3)129 (11.6)356 (11.3)184 (11.7)353 (9.7)171 (9.4)227 (7.6)102 (6.8)52 (5.1)22 (4.3)
> 100000101 (3.6)44 (3.1)67 (3.0)25 (2.2)80 (2.5)34 (2.2)99 (2.7)42 (2.3)55 (1.8)17 (1.1)14 (1.4)5 (1.0)
Unknown212 (7.6)103 (7.3)128 (5.8)77 (6.9)169 (5.4)121 (7.7)182 (5.0)96 (5.3)178 (5.9)92 (6.1)57 (5.6)32 (6.3)
Not answered380 (13.5)157 (11.2)269 (12.1)108 (9.7)419 (13.3)161 (10.2)465 (12.8)220 (12.1)425 (14.2)173 (11.5)146 (14.3)65 (12.8)
Physical activity (MET-minutes/week)2287 ± 21622264 ± 22012281 ± 21532223 ± 21762205 ± 19592197 ± 21422350 ± 22042366 ± 22312475 ± 22302376 ± 21882553 ± 21132523 ± 2360
Alcohol consumption
Never365 (13.0)209 (14.9)250 (11.2)142 (12.8)296 (9.4)148 (9.4)272 (7.5)121 (6.6)191 (6.4)92 (6.1)73 (7.2)33 (6.5)
Previous196 (7.0)125 (8.9)141 (6.3)96 (8.6)201 (6.4)129 (8.2)216 (5.9)130 (7.1)141 (4.7)116 (7.7)61 (6.0)33 (6.5)
Current2192 (78.1)1063 (75.8)1804 (81.1)868 (78.1)2623 (83.5)1288 (82.0)3144 (86.3)1563 (85.8)2662 (88.7)1292 (86.1)881 (86.5)442 (86.8)
Missing53 (1.9)6 (0.4)29 (1.3)6 (0.5)22 (0.7)6 (0.4)10 (0.3)7 (0.4)8 (0.3)1 (0.1)3 (0.3)1 (0.2)
Smoking
Never1549 (55.2)831 (59.2)1082 (48.7)554 (49.8)1524 (48.5)741 (47.2)1605 (44.1)779 (42.8)1261 (42.0)627 (41.8)406 (39.9)191 (37.5)
Former846 (30.1)366 (26.1)783 (35.2)380 (34.2)1245 (39.6)639 (40.7)1621 (44.5)840 (46.1)1472 (49.0)728 (48.5)511 (50.2)274 (53.8)
Current368 (13.1)193 (13.8)341 (15.3)173 (15.6)355 (11.3)176 (11.2)400 (11.0)188 (10.3)251 (8.4)133 (8.9)99 (9.7)41 (8.1)
Missing43 (1.5)13 (0.9)18 (0.8)5 (0.4)18 (0.6)15 (1.0)16 (0.4)14 (0.8)18 (0.6)13 (0.9)2 (0.2)3 (0.6)
Sleep duration (h)
< 7871 (31.0)462 (32.9)685 (30.8)391 (35.2)907 (28.9)475 (30.2)957 (26.3)517 (28.4)705 (23.5)359 (23.9)249 (24.5)124 (24.4)
7-91768 (63.0)845 (60.2)1436 (64.6)653 (58.7)2076 (66.1)1002 (63.8)2533 (69.5)1205 (66.2)2151 (71.7)1071 (71.4)727 (71.4)358 (70.3)
> 9107 (3.8)72 (5.1)57 (2.6)52 (4.7)106 (3.4)70 (4.5)116 (3.2)80 (4.4)112 (3.7)56 (3.7)31 (3.0)18 (3.5)
Missing60 (2.1)24 (1.7)46 (2.1)16 (1.4)53 (1.7)24 (1.5)36 (1.0)19 (1.0)34 (1.1)15 (1.0)11 (1.1)9 (1.8)
BMI (kg/m2)31.4 ± 6.831.5 ± 6.531.7 ± 6.631.9 ± 6.231.6 ± 6.431.8 ± 5.931.1 ± 5.731.2 ± 5.430.6 ± 5.430.8 ± 5.230.0 ± 4.830.2 ± 4.8
Cholesterol (mmol/L)4.70 ± 1.024.63 ± 1.104.64 ± 0.984.60 ± 1.094.63 ± 0.964.57 ± 1.074.59 ± 0.974.52 ± 1.034.65 ± 0.964.60 ± 1.044.76 ± 1.014.66 ± 1.05
HDL-C (mmol/L)1.22 ± 0.321.21 ± 0.321.21 ± 0.301.21 ± 0.321.23 ± 0.311.23 ± 0.301.23 ± 0.311.22 ± 0.301.25 ± 0.311.25 ± 0.301.28 ± 0.311.26 ± 0.31
LDL-C (mmol/L)2.85 ± 0.732.83 ± 0.822.80 ± 0.712.79 ± 0.802.77 ± 0.702.76 ± 0.782.73 ± 0.692.72 ± 0.752.78 ± 0.702.77 ± 0.782.86 ± 0.732.80 ± 0.77
Triglycerides (mmol/L)2.03 ± 1.432.02 ± 1.222.12 ± 1.472.10 ± 1.242.12 ± 1.432.09 ± 1.202.15 ± 1.382.16 ± 1.232.06 ± 1.272.08 ± 1.062.04 ± 1.202.07 ± 1.09
HbA1c (mmol/mol)36.3 ± 5.454.3 ± 17.2236.7 ± 5.852.6 ± 15.5236.9 ± 5.551.4 ± 14.2237.3 ± 5.349.7 ± 12.4237.3 ± 6.747.6 ± 10.8237.5 ± 4.546.5 ± 10.42
Hypertension1501 (53.5)764 (54.5)1307 (58.8)671 (60.3)2079 (66.2)1012 (64.4)2414 (66.3)1179 (64.7)1963 (65.4)984 (65.6)643 (63.2)340 (66.8)
Heart disease301 (10.7)147 (10.5)306 (13.8)154 (13.8)450 (14.3)221 (14.1)607 (16.7)293 (16.1)556 (18.5)271 (18.1)183 (18.0)90 (17.7)
Depression300 (10.7)142 (10.1)167 (7.5)99 (8.9)230 (7.3)109 (6.9)284 (7.8)142 (7.8)160 (5.3)82 (5.5)42 (4.1)22 (4.3)
1Age at diagnosis of type 2 diabetes.
2Refers to significant difference between diabetes participants and controls. T-test was used to test the difference of continuous variables between diabetes participants and controls and Chi-square for categorical variables.
Data are mean ± SD, or n (%). BMI: Body mass index; HbA1c: Glycosylated haemoglobin; HDL-C: High-density lipoprotein cholesterol; LDL-C: Low-density lipoprotein cholesterol; MET: Metabolic equivalent.

For vision acuity analysis, 20067 participants (37.8% females) aged 40-70 years (mean ±S D: 59.9 ± 7.9), were included. Diabetic participants across all age groups of diabetes diagnosis had higher HbA1c than the controls (Table 2). Individuals with T1D were more likely to have a normal sleep duration and higher HbA1c compared to the controls (Supplementary Table 3).

Table 2 Baseline characteristics of participants by diabetes and controls for analysis of vision acuity1.
< 45 yr
45-49 yr
50-54 yr
55-59 yr
60-64 yr
≥ 65 yr
Non-diabetes
Diabetes
Non-diabetes
Diabetes
Non-diabetes
Diabetes
Non-diabetes
Diabetes
Non-diabetes
Diabetes
Non-diabetes
Diabetes
Age (yr)53.1 ± 8.652.8 ± 8.554.8 ± 8.554.7 ± 5.859.0 ± 7.958.9 ± 4.862.5 ± 5.962.3 ± 3.565.2 ± 3.665.0 ± 2.467.5 ± 2.867.6 ± 1.3
Gender
Female751 (37.6)379 (37.9)577 (36.4)284 (35.8)937 (39.2)480 (40.2)1035 (35.5)512 (35.1)962 (36.7)485 (37.1)363 (41.4)177 (40.4)
Male1249 (62.5)621 (62.1)1009 (63.6)509 (64.2)1453 (60.8)715 (59.8)1879 (64.5)945 (64.9)1656 (63.3)824 (62.9)513 (58.6)261 (59.6)
Ethnicity
Whites1268 (63.4)606 (60.6)1078 (68.0)518 (65.3)1793 (75.0)866 (72.5)2478 (85.0)1223 (83.9)2302 (87.9)1122 (85.7)768 (87.7)379 (86.5)
Non-whites611 (30.6)373 (37.3)454 (28.6)270 (34.0)511 (21.4)306 (25.6)378 (13.0)222 (15.2)267 (10.2)172 (13.1)86 (9.8)54 (12.3)
Unknown121 (6.1)21 (2.1)54 (3.4)5 (0.6)86 (3.6)23 (1.9)58 (2.0)12 (0.8)49 (1.9)15 (1.1)22 (2.5)5 (1.1)
Education
0-5 yr585 (29.3)285 (28.5)490 (30.9)215 (27.1)683 (28.6)330 (27.6)755 (25.9)359 (24.6)619 (23.6)306 (23.4)193 (22.0)92 (21.0)
6-12 yr942 (47.1)487 (48.7)811 (51.1)430 (54.2)1171 (49.0)589 (49.3)1359 (46.6)693 (47.6)1156 (44.2)590 (45.1)375 (42.8)183 (41.8)
≥ 13 yr369 (18.5)192 (19.2)239 (15.1)130 (16.4)462 (19.3)244 (20.4)736 (25.3)370 (25.4)783 (29.9)385 (29.4)287 (32.8)151 (34.5)
Missing104 (5.2)36 (3.6)46 (2.9)18 (2.3)74 (3.1)32 (2.7)64 (2.2)35 (2.4)60 (2.3)28 (2.1)21 (2.4)12 (2.7)
Household income (pounds)
< 18000475 (23.8)266 (26.6)359 (22.6)217 (27.4)607 (25.4)330 (27.6)842 (28.9)418 (28.7)796 (30.4)414 (31.6)286 (32.6)165 (37.7)
18000-30999377 (18.9)212 (21.2)346 (21.8)167 (21.1)521 (21.8)244 (20.4)718 (24.6)370 (25.4)673 (25.7)341 (26.1)227 (25.9)106 (24.2)
31000-51999372 (18.6)179 (17.9)313 (19.7)143 (18.0)415 (17.4)205 (17.2)493 (16.9)245 (16.8)402 (15.4)199 (15.2)120 (13.7)55 (12.6)
52000-100000256 (12.8)113 (11.3)234 (14.8)106 (13.4)298 (12.5)151 (12.6)301 (10.3)149 (10.2)220 (8.4)106 (8.1)52 (5.9)23 (5.3)
> 10000072 (3.6)34 (3.4)61 (3.8)24 (3.0)77 (3.2)29 (2.4)82 (2.8)36 (2.5)54 (2.1)15 (1.1)14 (1.6)5 (1.1)
Unknown132 (6.6)76 (7.6)70 (4.4)56 (7.1)122 (5.1)83 (6.9)117 (4.0)71 (4.9)126 (4.8)71 (5.4)44 (5.0)24 (5.5)
Not answered316 (15.8)120 (12.0)203 (12.8)80 (10.1)350 (14.6)153 (12.8)361 (12.4)168 (11.5)347 (13.3)163 (12.5)133 (15.2)60 (13.7)
Physical activity (MET-minutes/week)2225 ± 20872269 ± 21342154 ± 20192147 ± 20862206 ± 20232180 ± 21712433 ± 21902384 ± 22042410 ± 22062426 ± 22312591 ± 23412518 ± 2271
Alcohol consumption
Never233 (11.7)156 (15.6)178 (11.2)106 (13.4)205 (8.6)135 (11.3)213 (7.3)107 (7.3)168 (6.4)91 (7.0)60 (6.8)30 (6.8)
Previous110 (5.5)78 (7.8)96 (6.1)58 (7.3)138 (5.8)85 (7.1)170 (5.8)81 (5.6)131 (5.0)77 (5.9)55 (6.3)34 (7.8)
Current1595 (79.8)755 (75.5)1283 (80.9)623 (78.6)2000 (83.7)966 (80.8)2509 (86.1)1260 (86.5)2294 (87.6)1133 (86.6)751 (85.7)371 (84.7)
Missing62 (3.1)11 (1.1)29 (1.8)6 (0.8)47 (2.0)9 (0.8)22 (0.8)9 (0.6)25 (1.0)8 (0.6)10 (1.1)3 (0.7)
Smoking
Never1089 (54.5)591 (59.1)778 (49.1)405 (51.1)1132 (47.4)590 (49.4)1271 (43.6)648 (44.5)1144 (43.7)550 (42.0)383 (43.7)182 (41.6)
Former581 (29.1)260 (26.0)563 (35.5)258 (32.5)946 (39.6)458 (38.3)1308 (44.9)648 (44.5)1220 (46.6)632 (48.3)410 (46.8)222 (50.7)
Current260 (13.0)130 (13.0)222 (14.0)124 (15.6)273 (11.4)130 (10.9)312 (10.7)145 (10.0)221 (8.4)109 (8.3)70 (8.0)31 (7.1)
Missing70 (3.5)19 (1.9)23 (1.5)6 (0.8)39 (1.6)17 (1.4)23 (0.8)16 (1.1)33 (1.3)18 (1.4)13 (1.5)3 (0.7)
Sleep duration (h)
< 7580 (29.0)313 (31.3)474 (29.9)281 (35.4)652 (27.3)347 (29.0)745 (25.6)392 (26.9)594 (22.7)314 (24.0)231 (26.4)98 (22.4)
7-91274 (63.7)625 (62.5)1025 (64.6)461 (58.1)1604 (67.1)782 (65.4)2043 (70.1)982 (67.4)1880 (71.8)923 (70.5)610 (69.6)320 (73.1)
> 964 (3.2)39 (3.9)48 (3.0)37 (4.7)72 (3.0)42 (3.5)90 (3.1)63 (4.3)102 (3.9)55 (4.2)23 (2.6)11 (2.5)
Missing82 (4.1)23 (2.3)39 (2.5)14 (1.8)62 (2.6)24 (2.0)36 (1.2)20 (1.4)42 (1.6)17 (1.3)12 (1.4)9 (2.1)
BMI (kg/m2)31.3 ± 6.731.2 ± 6.731.8 ± 6.632.0 ± 6.331.0 ± 6.131.3 ± 5.630.8 ± 5.530.9 ± 5.130.6 ± 5.530.6 ± 5.129.8 ± 4.930.1 ± 4.9
Cholesterol (mmol/L)4.67 ± 1.014.62 ± 1.114.63 ± 1.004.59 ± 1.084.59 ± 0.954.58 ± 1.064.58 ± 0.954.52 ± 1.054.68 ± 0.954.61 ± 1.044.71 ± 0.964.68 ± 1.06
HDL-C (mmol/L)1.24 ± 0.311.23 ± 0.331.21 ± 0.311.22 ± 0.331.24 ± 0.301.24 ± 0.311.24 ± 0.311.24 ± 0.301.27 ± 0.301.26 ± 0.311.29 ± 0.321.27 ± 0.31
LDL-C (mmol/L)2.83 ± 0.752.81 ± 0.832.79 ± 0.722.78 ± 0.782.75 ± 0.682.76 ± 0.782.72 ± 0.682.70 ± 0.762.80 ± 0.692.76 ± 0.772.83 ± 0.712.82 ± 0.78
Triglycerides (mmol/L)1.88 ± 1.281.92 ± 1.152.08 ± 1.472.07 ± 1.222.01 ± 1.372.00 ± 1.102.11 ± 1.412.10 ± 1.212.01 ± 1.212.02 ± 1.041.96 ± 1.191.98 ± 1.01
HbA1c (mmol/mol)36.6 ± 5.353.5 ± 16.6236.6 ± 5.052.8 ± 15.6237.0 ± 4.951.2 ± 13.8237.1 ± 5.349.4 ± 12.5237.3 ± 7.047.4 ± 10.9237.6 ± 10.345.5 ± 9.62
Hypertension1058 (52.9)528 (52.8)971 (61.2)480 (60.5)1486 (62.2)746 (62.4)1883 (64.6)930 (63.8)1707 (65.2)842 (64.3)547 (62.4)280 (63.9)
Heart disease200 (10.0)99 (9.9)177 (11.2)95 (12.0)314 (13.1)157 (13.1)481 (16.5)230 (15.8)447 (17.1)222 (17.0)158 (18.0)78 (17.8)
Depression162 (8.1)84 (8.4)129 (8.1)66 (8.3)146 (6.1)70 (5.9)190 (6.5)95 (6.5)127 (4.9)63 (4.8)24 (2.7)12 (2.7)
1Age at diagnosis of type 2 diabetes.
2Refers to significant difference between diabetes participants and controls. T-test was used to test the difference of continuous variables between diabetes participants and controls and Chi-square for categorical variables.
Data are means ± SD, or n (%). BMI: Body mass index; HbA1c: Glycosylated haemoglobin; MET: Metabolic equivalent.
Incidence of ocular conditions

Over a median follow-up of 11.0 years (interquartile range: 10.7-11.5), 3874 new cases of cataract, 665 new cases of glaucoma, and 616 new cases of AMD were identified.

Age at diagnosis of diabetes and incident AMD, cataract, and glaucoma

As shown in Figure 2, the relative risk for incident AMD associated with diabetes decreased with the increasing age at diagnosis of diabetes. In the multivariable-adjusted analysis, T2D diagnosed at age of < 45 [HR (95%CI): 2.71 (1.49-4.93)], 45-49 [2.57 (1.17-5.65)], 50-54 [1.85 (1.13-3.04)], or 55-59 years [1.53 (1.00-2.34)] was associated with a higher risk of incident AMD. T1D [HR (95%CI): 4.12 (1.99-8.53)] was associated with an increased risk of AMD independent of concurrent HbA1c.

Figure 2
Open in New Tab   Full Size Figure   Download Figure
Figure 2 Risk for ocular conditions associated with age at diagnosis of diabetes. Cox proportional hazard regression models were used to examine the association between diabetes and incident ocular condition for each group of diabetes diagnosis age. Model 1 was adjusted for age, gender, ethnicity, income, education, alcohol consumption, physical activity, sleep duration, smoking, body mass index, depression, hypertension, heart disease, stroke, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglyceride. Central squares of each horizontal line represent the hazard ratio for each subgroup. Horizontal lines indicate the range of the 95%CI. The vertical dash lines indicate the hazard ratio of 1.0. AMD: Age-related macular degeneration; T1D: Type 1 diabetes; T2D: Type 2 diabetes; HbA1C: Glycated haemoglobin.

Similarly, the association between diabetes and glaucoma was dependent on the age at diagnosis of diabetes. After adjustment for HbA1c and other covariates, only diabetes diagnosed at age of < 45 years only [HR (95%CI): 1.76 (1.00-3.12)] was associated with an increased risk of glaucoma. The multivariable-adjusted HR (95%CI) for glaucoma associated with T1D was 2.40 (1.09-5.31).

In the multivariable-adjusted model, the HRs (95%CIs) for incident cataract associated with diabetes diagnosed at < 45, 45-49, 50-54, and 55-59 years of age were 2.18 (1.71-2.79), 1.54 (1.19-2.01), 1.60 (1.31-1.96), and 1.21 (1.02-1.43), respectively. T1D was independently associated with an increased risk of incident cataract [2.95 (2.17-4.02)].

As shown in Supplementary Figure 2, T2D diagnosed at < 45, 45-49, 50-54, and 55-59, but not 60-64 or ≥ 65 years of age was associated with an increased risk of cataract surgery, where individuals with T2D diagnosed < 45 years had the highest excess risk of cataract surgery [HR (95%CI): 2.67 (1.88-3.79)]. The multivariable-adjusted HR (95%CI) for cataract surgery associated with T1D was 4.63 (3.10-6.93).

Age at diagnosis of diabetes and vision acuity

After adjustment for covariates and HbA1c, individuals with T2D diagnosed at age of < 45 [β 95%CI: 0.025 (0.009, 0.040)], and 50-54 years [0.016 (0.002, 0.029)] had higher LogMAR compared to the corresponding controls. T1D was associated with a larger LogMAR [0.044 (0.015,0.073), Figure 3].

Figure 3
Open in New Tab   Full Size Figure   Download Figure
Figure 3 Vision acuity associated with age at diagnosis of diabetes. General linear regression models were used to test the difference in LogMAR between diabetic participants and controls for each group of diabetes diagnosis age. Model 1 was adjusted for age, gender, ethnicity, income, education, alcohol consumption, physical activity, sleep duration, smoking, BMI, depression, hypertension, heart disease, stroke, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglyceride. Central squares of each horizontal line represent the β for each subgroup. Horizontal lines indicate the range of the 95%CI. The vertical dash lines represent the β of 0. T1D: Type 1 diabetes; T2D: Type 2 diabetes; HbA1C: Glycated haemoglobin.
Age at diagnosis of diabetes and IOP

As shown in Figure 4, individuals with T2D diagnosed at < 45 [β (95%CI): 0.88 (0.59, 1.18) mmHg], 45-49 [0.86 (0.53, 1.18) mmHg], and 50-54 years of age [0.78 (0.52, 1.05) mmHg] had higher IOP compared with the controls. The β (95%CI) for IOP associated with T1D was larger [1.15 (0.73,1.56) mmHg].

Figure 4
Open in New Tab   Full Size Figure   Download Figure
Figure 4 Intraocular pressure associated with age at diagnosis of diabetes. General linear regression models were used to test the difference in intraocular pressure between diabetic participants and controls for each group of diabetes diagnosis age. Model 1 was adjusted for age, gender, ethnicity, income, education, alcohol consumption, physical activity, sleep duration, smoking, body mass index, depression, hypertension, heart disease, stroke, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglyceride. T2D: Type 2 diabetes; IOP: Intraocular pressure.
Sensitivity analysis

Individuals with diabetes diagnosed at < 50 years of age were younger but had a higher incidence of ocular diseases compared with controls (Supplementary Figure 3). A larger HR was observed for those with diabetes diagnosed at older age. After adjustment for covariates, the association was reversed with diabetes diagnosed at younger age associated with a larger HR. This trend remained consistent after further adjustment for diabetes duration (Figure 5). Individuals with diabetes diagnosed at < 45, 45-49, or 50-54 years of age were younger and had higher LogMAR compared with controls (Supplementary Figure 4). Older age at the diagnosis of diabetes was associated with a larger increase in LogMAR compared with controls. However, after adjustment for covariates, diabetes diagnosed at a younger age was associated with a larger increase in LogMAR (Supplementary Figure 5).

Figure 5
Open in New Tab   Full Size Figure   Download Figure
Figure 5 Risk for ocular conditions associated with age at diagnosis of diabetes with the same reference. Sensitivity analysis was conducted to randomly select controls for each individual with type 2 diabetes with all diabetic patients as a whole. Cox proportional hazard regression models were used to estimate hazard ratios for ocular conditions associated with age at diagnosis of diabetes with controls as the reference for each group of diabetes diagnosed age. The multivariable model was adjusted for age, gender, ethnicity, income, education, alcohol consumption, physical activity, sleep duration, smoking, body mass index, depression, hypertension, heart disease, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglyceride, and glycated haemoglobin. Central squares of each horizontal line represent the hazard ratio for each subgroup. Horizontal lines indicate the range of the 95%CI. The vertical dash lines indicate the hazard ratio of 1.0. T2D: Type 2 diabetes; AMD: Age-related macular degeneration.
DISCUSSION

This large prospective cohort study demonstrated that younger age at diagnosis of diabetes was associated with a larger relative risk for cataract, glaucoma, and AMD independent of concurrent HbA1c levels. Individuals with T2D diagnosed before the age of 45 years were more than twice as likely to develop these ocular conditions, while those with T1D exhibited a more pronounced relative risk. Similarly, T2D diagnosed before the age of 55 years and T1D were associated with an increased LogMAR. Sensitivity analysis suggests these associations are independent of duration of diabetes.

Diabetes is one of the most important determinants for cataract[8,28,29]. We found that diabetes was associated with an increased risk of incident cataract, and in particular diabetes diagnosed at < 45 years of age had larger excessive risk of cataract. To our knowledge, no previous study has investigated the impact of age at diagnosis of diabetes on the association between diabetes and cataract. However, several studies have shown that the association between diabetes and cataract was stronger among younger than older adults[28-30]. In a cross-sectional analysis, longer duration of diabetes was associated with a higher prevalence of cataract[28]. These studies provide indirect evidence for the rationale of our findings that younger age at diagnosis of diabetes was associated with a larger excess risk of cataract.

Previous studies have been inconsistent on the association between diabetes and glaucoma. Although a meta-analysis showed that diabetes was associated with a higher risk of glaucoma [relative risk (95%CI): 1.36 (1.25-1.50)], only three out of seven prospective studies included in the meta-analysis found a significant association between diabetes and glaucoma[11]. The lack of significance in some studies may be attributed to a relatively short duration of diabetes[31]. We found individuals with T1D or T2D diagnosed before the age of 45 years but not at 45 years or older had a higher risk of glaucoma. This is consistent with previous studies showing controversial associations between diabetes and glaucoma. It is possible that cumulative exposure to hyperglycemia from an early life may contribute to increased IOP[32], thus elevating the risk of glaucoma. This is supportive by further analysis demonstrating that diabetes diagnosed at a younger age was associated with a larger increase in IOP.

A meta-analysis showed that diabetes was associated with an increased risk of incident AMD [relative risk (95%CI): 1.05 (1.00–1.11)], although the effect size is small[19]. Among 7 cohort studies in this meta-analysis, only one study reported a significant association[33]. Another prospective study (not included in this meta-analysis) of 71904 patients with diabetes and 270213 patients without diabetes found no significant association between diabetes and incident AMD[34]. A recent prospective study even found that diabetes was associated with a decreased progression of AMD[35]. However, we found that diabetes diagnosed at a younger age but not at an older age was associated with an increased risk of AMD. This finding may offer an explanation for the lack of significant associations reported in most previous studies. Notably, previous studies often combined individuals with diabetes diagnosed at both younger and older ages, which may introduce a bias towards a null association.

Whether the association between diabetes and incident cataract, glaucoma, or AMD is moderated by the age at diagnosis of diabetes has not been reported in previous studies. However, our study is consistent with a cross-sectional study of 3322 individuals demonstrating that early-onset T2D was associated with a higher prevalence of diabetic retinopathy[36]. Likewise, diabetes diagnosed at a younger age was associated with a larger excess risk of cardiovascular disease and mortality[20]. Our further analysis showed that diabetes diagnosed at < 50 years, but not ≥ 50 years of age was associated with decreased vision. The potential effect of diabetes on the development of ocular conditions and vision loss is independent of HbA1c, highlighting the importance of the age at diabetes diagnosis rather than management of diabetes in this association. T1D was associated with a larger effect size for ocular conditions compared with T2D, even when diagnosed at a younger age. The more potentially harmful effect of T1D may stem from its longer duration of hyperglycaemia and insulin dependency, leading to more extensive damage to blood vessels and nervous system.

The mechanisms undelying the association between a younger age at the diagnosis of diabetes and ocular conditions and vision loss remain largely unknown. A prospective study has shown that T2D developed at a younger age was associated with a higher risk of obesity, worse lipid profiles and higher HbA1c, and a faster deterioration in glycaemic control compared to those with diabetes onset at an older age[37]. These markers have been shown to be important determinants for cataract[38] and glaucoma[31,39] among diabetic patients. This may indicate that early-onset diabetes may represent a more pathogenic condition than late-onset disease for the development of ocular conditions[37]. Furthermore, cumulative exposure to diabetes from early to middle adulthood may exert substantial adverse effects on the development of ocular conditions[40]. The stronger association between early-onset diabetes and ocular conditions may also be attributed to the shared genetics between diabetes and ocular conditions[41-43]. However, the clear pathogenesis of ocular conditions, especially AMD due to diabetes, needs further exploration in research.

To the best of our knowledge, this is the first prospective cohort study to examine the association of age at the diagnosis of diabetes with main ocular conditions. There are several potential limitations in our study. Firstly, some cases of incident ocular conditions may not be captured using inpatient data. Secondly, it is possible that some controls may have developed diabetes during follow-up, which is more likely to bias the associations towards the null. Thirdly, a large proportion of the individuals without eye health data in the UK biobank cohort were excluded from the analysis. This may limit the generalizability of our findings to the whole population.

CONCLUSION

In conclusion, our findings suggest the age at the diagnosis of diabetes plays an important role in the association between diabetes and incident cataract, glaucoma, and AMD as well as vision. A younger age at the diagnosis of diabetes was associated with larger excessive relative risk for ocular conditions and larger vision loss. T1D appears to have potentially more harmful effects.

ARTICLE HIGHLIGHTS
Research background

Diabetes has been linked to numerous ocular conditions, including cataract, glaucomaand age-related macular degeneration (AMD). Several studies have demonstrated a positive relationship between diabetes and AMD, but more studies did not find a significant association. Diabetes may have different associations with different stages of ocular conditions, and the duration of diabetes may affect the development of diabetic eye disease. It is important to identify the life stage at which a diagnosis of diabetes is associated with the highest risk of najor ocular conditions for the prevention or screening of these conditions.

Research motivation

To examine associations between the age of diabetes diagnosis and the incidence of cataract, glaucoma, AMD, and vision acuity. It is important to identify the life stage at which a diagnosis of diabetes is associated with the highest risk of najor ocular conditions for the prevention or screening of these conditions.

Research objectives

To examine associations between the age of diabetes diagnosis and the incidence of cataract, glaucoma, AMD, and vision acuity. A stronger association between diabetes and incident ocular conditions was observed where diabetes was diagnosed at a younger age. It is important to identify the life stage at which a diagnosis of diabetes is associated with the highest risk of major ocular conditions for the prevention or screening of these conditions, and the clear pathogenesis of ocular conditions, needs further exploration in research.

Research methods

This is the first prospective cohort study to examine the association of age at the diagnosis of diabetes with main ocular conditions. Our analysis was using the UK Biobank. The cohort included 8709 diabetic participants and 17418 controls for ocular condition analysis, and 6689 diabetic participants and 13378 controls for vision analysis. Ocular diseases were identified using inpatient records until January 2021. Vision acuity was assessed using a chart.

Research results

This large prospective cohort study demonstrated that younger age at diagnosis of diabetes was associated with a larger relative risk for cataract, glaucoma, and AMD independent of concurrent glycated haemoglobin levels. Individuals with type 2 diabetes (T2D) diagnosed before the age of 45 years were more than twice as likely to develop these ocular conditions, while those with type 1 diabetes (T1D) exhibited a more pronounced relative risk. Similarly, T2D diagnosed before the age of 55 years and T1D were associated with an increased LogMAR. However, the clear pathogenesis of ocular conditions, especially AMD due to diabetes, needs further exploration in research.

Research conclusions

Our findings suggest the age at the diagnosis of diabetes plays an important role in the association between diabetes and incident cataract, glaucoma, and AMD as well as vision. A younger age at the diagnosis of diabetes was associated with larger excessive relative risk for ocular conditions and larger vision loss. T1D appears to have potentially more harmful effects.

Research perspectives

Investigated the impact of age at diagnosis of diabetes on the association between diabetes and cataract, glaucoma, AMD, and vision acuity, by the more detailed breakdown of factors. To analyse more about the shared genetics between diabetes and ocular conditions.

ACKNOWLEDGEMENTS

This research has been conducted using the UK Biobank Resource under Application Number (62443, 62525, 62491, 94372, 105658). We thank the participants of the UK Biobank. We thank the language proofreading by Shahin Alam.

Footnotes

Provenance and peer review: Unsolicited article; Externally peer reviewed.

Peer-review model: Single blind

Specialty type: Endocrinology and metabolism

Country/Territory of origin: China

Peer-review report’s scientific quality classification

Grade A (Excellent): A

Grade B (Very good): B

Grade C (Good): 0

Grade D (Fair): 0

Grade E (Poor): 0

P-Reviewer: Al-Suhaimi EA, Saudi Arabia; Cai L, United States S-Editor: Lin C L-Editor: A P-Editor: Zhang YL

References
1.  GBD 2019 Blindness and Vision Impairment Collaborators; Vision Loss Expert Group of the Global Burden of Disease Study. Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study. Lancet Glob Health. 2021;9:e144-e160.  [PubMed]  [DOI]  [Cited in This Article: ]
2.  Keel S, Cieza A. Rising to the challenge: estimates of the magnitude and causes of vision impairment and blindness. Lancet Glob Health. 2021;9:e100-e101.  [PubMed]  [DOI]  [Cited in This Article: ]
3.  Pearce I, Simó R, Lövestam-Adrian M, Wong DT, Evans M. Association between diabetic eye disease and other complications of diabetes: Implications for care. A systematic review. Diabetes Obes Metab. 2019;21:467-478.  [PubMed]  [DOI]  [Cited in This Article: ]
4.  Araújo LR, Orefice JL, Gonçalves MA, Guimarães NS, Soares AN, Salomon T, de Souza AH. Use of digital retinography to detect vascular changes in pre-diabetic patients: a cross-sectional study. Diabetol Metab Syndr. 2023;15:225.  [PubMed]  [DOI]  [Cited in This Article: ]
5.  Kiziltoprak H, Tekin K, Inanc M, Goker YS. Cataract in diabetes mellitus. World J Diabetes. 2019;10:140-153.  [PubMed]  [DOI]  [Cited in This Article: ]
6.  Fujita A, Hashimoto Y, Matsui H, Yasunaga H, Aihara M. Association between lifestyle habits and glaucoma incidence: a retrospective cohort study. Eye (Lond). 2023;37:3470-3476.  [PubMed]  [DOI]  [Cited in This Article: ]
7.  Heesterbeek TJ, Lorés-Motta L, Hoyng CB, Lechanteur YTE, den Hollander AI. Risk factors for progression of age-related macular degeneration. Ophthalmic Physiol Opt. 2020;40:140-170.  [PubMed]  [DOI]  [Cited in This Article: ]
8.  Floud S, Kuper H, Reeves GK, Beral V, Green J. Risk Factors for Cataracts Treated Surgically in Postmenopausal Women. Ophthalmology. 2016;123:1704-1710.  [PubMed]  [DOI]  [Cited in This Article: ]
9.  Chan TCW, Bala C, Siu A, Wan F, White A. Risk Factors for Rapid Glaucoma Disease Progression. Am J Ophthalmol. 2017;180:151-157.  [PubMed]  [DOI]  [Cited in This Article: ]
10.  Jiang X, Varma R, Wu S, Torres M, Azen SP, Francis BA, Chopra V, Nguyen BB; Los Angeles Latino Eye Study Group. Baseline risk factors that predict the development of open-angle glaucoma in a population: the Los Angeles Latino Eye Study. Ophthalmology. 2012;119:2245-2253.  [PubMed]  [DOI]  [Cited in This Article: ]
11.  Zhao YX, Chen XW. Diabetes and risk of glaucoma: systematic review and a Meta-analysis of prospective cohort studies. Int J Ophthalmol. 2017;10:1430-1435.  [PubMed]  [DOI]  [Cited in This Article: ]
12.  Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FL 3rd; Age-Related Eye Disease Study Research Group. Risk factors for the incidence of Advanced Age-Related Macular Degeneration in the Age-Related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology. 2005;112:533-539.  [PubMed]  [DOI]  [Cited in This Article: ]
13.  Wang IK, Lin HJ, Wan L, Lin CL, Yen TH, Sung FC. Risk of age-related macular degeneration in end-stage renal disease patients receiving long-term dialysis. Retina. 2016;36:1866-1873.  [PubMed]  [DOI]  [Cited in This Article: ]
14.  Tomany SC, Wang JJ, Van Leeuwen R, Klein R, Mitchell P, Vingerling JR, Klein BE, Smith W, De Jong PT. Risk factors for incident age-related macular degeneration: pooled findings from 3 continents. Ophthalmology. 2004;111:1280-1287.  [PubMed]  [DOI]  [Cited in This Article: ]
15.  Jonasson F, Fisher DE, Eiriksdottir G, Sigurdsson S, Klein R, Launer LJ, Harris T, Gudnason V, Cotch MF. Five-year incidence, progression, and risk factors for age-related macular degeneration: the age, gene/environment susceptibility study. Ophthalmology. 2014;121:1766-1772.  [PubMed]  [DOI]  [Cited in This Article: ]
16.  Saunier V, Merle BMJ, Delyfer MN, Cougnard-Grégoire A, Rougier MB, Amouyel P, Lambert JC, Dartigues JF, Korobelnik JF, Delcourt C. Incidence of and Risk Factors Associated With Age-Related Macular Degeneration: Four-Year Follow-up From the ALIENOR Study. JAMA Ophthalmol. 2018;136:473-481.  [PubMed]  [DOI]  [Cited in This Article: ]
17.  Buch H, Vinding T, la Cour M, Jensen GB, Prause JU, Nielsen NV. Risk factors for age-related maculopathy in a 14-year follow-up study: the Copenhagen City Eye Study. Acta Ophthalmol Scand. 2005;83:409-418.  [PubMed]  [DOI]  [Cited in This Article: ]
18.  Tan JS, Mitchell P, Smith W, Wang JJ. Cardiovascular risk factors and the long-term incidence of age-related macular degeneration: the Blue Mountains Eye Study. Ophthalmology. 2007;114:1143-1150.  [PubMed]  [DOI]  [Cited in This Article: ]
19.  Chen X, Rong SS, Xu Q, Tang FY, Liu Y, Gu H, Tam PO, Chen LJ, Brelén ME, Pang CP, Zhao C. Diabetes mellitus and risk of age-related macular degeneration: a systematic review and meta-analysis. PLoS One. 2014;9:e108196.  [PubMed]  [DOI]  [Cited in This Article: ]
20.  Sattar N, Rawshani A, Franzén S, Svensson AM, Rosengren A, McGuire DK, Eliasson B, Gudbjörnsdottir S. Age at Diagnosis of Type 2 Diabetes Mellitus and Associations With Cardiovascular and Mortality Risks. Circulation. 2019;139:2228-2237.  [PubMed]  [DOI]  [Cited in This Article: ]
21.  Ramsey DJ, Kwan JT, Sharma A. Keeping an eye on the diabetic foot: The connection between diabetic eye disease and wound healing in the lower extremity. World J Diabetes. 2022;13:1035-1048.  [PubMed]  [DOI]  [Cited in This Article: ]
22.  Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, Downey P, Elliott P, Green J, Landray M, Liu B, Matthews P, Ong G, Pell J, Silman A, Young A, Sprosen T, Peakman T, Collins R. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12:e1001779.  [PubMed]  [DOI]  [Cited in This Article: ]
23.  Eastwood SV, Mathur R, Atkinson M, Brophy S, Sudlow C, Flaig R, de Lusignan S, Allen N, Chaturvedi N. Algorithms for the Capture and Adjudication of Prevalent and Incident Diabetes in UK Biobank. PLoS One. 2016;11:e0162388.  [PubMed]  [DOI]  [Cited in This Article: ]
24.  Chua SYL, Luben RN, Hayat S, Broadway DC, Khaw KT, Warwick A, Britten A, Day AC, Strouthidis N, Patel PJ, Khaw PT, Foster PJ, Khawaja AP; UK Biobank Eye and Vision Consortium. Alcohol Consumption and Incident Cataract Surgery in Two Large UK Cohorts. Ophthalmology. 2021;128:837-847.  [PubMed]  [DOI]  [Cited in This Article: ]
25.  Chua SYL, Thomas D, Allen N, Lotery A, Desai P, Patel P, Muthy Z, Sudlow C, Peto T, Khaw PT, Foster PJ; UK Biobank Eye & Vision Consortium. Cohort profile: design and methods in the eye and vision consortium of UK Biobank. BMJ Open. 2019;9:e025077.  [PubMed]  [DOI]  [Cited in This Article: ]
26.  International Physical Activity Questionnaire.   Guidelines for data processing and analysis of the International Physical Activity Questionnaire (IPAQ). Nov 2005 [cited 3 August 2022]. Available from: https://biobank.ndph.ox.ac.uk/showcase/ukb/docs/ipaq_analysis.pdf.  [PubMed]  [DOI]  [Cited in This Article: ]
27.  Ferrie JE, Shipley MJ, Cappuccio FP, Brunner E, Miller MA, Kumari M, Marmot MG. A prospective study of change in sleep duration: associations with mortality in the Whitehall II cohort. Sleep. 2007;30:1659-1666.  [PubMed]  [DOI]  [Cited in This Article: ]
28.  Klein BE, Klein R, Moss SE. Prevalence of cataracts in a population-based study of persons with diabetes mellitus. Ophthalmology. 1985;92:1191-1196.  [PubMed]  [DOI]  [Cited in This Article: ]
29.  Klein BE, Klein R, Moss SE. Incidence of cataract surgery in the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Am J Ophthalmol. 1995;119:295-300.  [PubMed]  [DOI]  [Cited in This Article: ]
30.  Nielsen NV, Vinding T. The prevalence of cataract in insulin-dependent and non-insulin-dependent-diabetes mellitus. Acta Ophthalmol (Copenh). 1984;62:595-602.  [PubMed]  [DOI]  [Cited in This Article: ]
31.  Zhao D, Cho J, Kim MH, Friedman DS, Guallar E. Diabetes, fasting glucose, and the risk of glaucoma: a meta-analysis. Ophthalmology. 2015;122:72-78.  [PubMed]  [DOI]  [Cited in This Article: ]
32.  Weinreb RN, Aung T, Medeiros FA. The pathophysiology and treatment of glaucoma: a review. JAMA. 2014;311:1901-1911.  [PubMed]  [DOI]  [Cited in This Article: ]
33.  Shalev V, Sror M, Goldshtein I, Kokia E, Chodick G. Statin use and the risk of age related macular degeneration in a large health organization in Israel. Ophthalmic Epidemiol. 2011;18:83-90.  [PubMed]  [DOI]  [Cited in This Article: ]
34.  He MS, Chang FL, Lin HZ, Wu JL, Hsieh TC, Lee YC. The Association Between Diabetes and Age-Related Macular Degeneration Among the Elderly in Taiwan. Diabetes Care. 2018;41:2202-2211.  [PubMed]  [DOI]  [Cited in This Article: ]
35.  Chakravarthy U, Bailey CC, Scanlon PH, McKibbin M, Khan RS, Mahmood S, Downey L, Dhingra N, Brand C, Brittain CJ, Willis JR, Venerus A, Muthutantri A, Cantrell RA. Progression from Early/Intermediate to Advanced Forms of Age-Related Macular Degeneration in a Large UK Cohort: Rates and Risk Factors. Ophthalmol Retina. 2020;4:662-672.  [PubMed]  [DOI]  [Cited in This Article: ]
36.  Middleton TL, Constantino MI, Molyneaux L, D'Souza M, Twigg SM, Wu T, Yue DK, Zoungas S, Wong J. Young-onset type 2 diabetes and younger current age: increased susceptibility to retinopathy in contrast to other complications. Diabet Med. 2020;37:991-999.  [PubMed]  [DOI]  [Cited in This Article: ]
37.  Steinarsson AO, Rawshani A, Gudbjörnsdottir S, Franzén S, Svensson AM, Sattar N. Short-term progression of cardiometabolic risk factors in relation to age at type 2 diabetes diagnosis: a longitudinal observational study of 100,606 individuals from the Swedish National Diabetes Register. Diabetologia. 2018;61:599-606.  [PubMed]  [DOI]  [Cited in This Article: ]
38.  Drinkwater JJ, Davis WA, Davis TME. A systematic review of risk factors for cataract in type 2 diabetes. Diabetes Metab Res Rev. 2019;35:e3073.  [PubMed]  [DOI]  [Cited in This Article: ]
39.  Song BJ, Aiello LP, Pasquale LR. Presence and Risk Factors for Glaucoma in Patients with Diabetes. Curr Diab Rep. 2016;16:124.  [PubMed]  [DOI]  [Cited in This Article: ]
40.  Cha AE, Villarroel MA, Vahratian A. Eye Disorders and Vision Loss Among U.S. Adults Aged 45 and Over With Diagnosed Diabetes, 2016-2017. NCHS Data Brief. 2019;1-8.  [PubMed]  [DOI]  [Cited in This Article: ]
41.  Shen L, Walter S, Melles RB, Glymour MM, Jorgenson E. Diabetes Pathology and Risk of Primary Open-Angle Glaucoma: Evaluating Causal Mechanisms by Using Genetic Information. Am J Epidemiol. 2016;183:147-155.  [PubMed]  [DOI]  [Cited in This Article: ]
42.  Chang C, Zhang K, Veluchamy A, Hébert HL, Looker HC, Colhoun HM, Palmer CN, Meng W. A Genome-Wide Association Study Provides New Evidence That CACNA1C Gene is Associated With Diabetic Cataract. Invest Ophthalmol Vis Sci. 2016;57:2246-2250.  [PubMed]  [DOI]  [Cited in This Article: ]
43.  Ludwig PE, Freeman SC, Janot AC. Novel stem cell and gene therapy in diabetic retinopathy, age related macular degeneration, and retinitis pigmentosa. Int J Retina Vitreous. 2019;5:7.  [PubMed]  [DOI]  [Cited in This Article: ]