Association Between Myopia, Ultraviolet B RadiationExposure, Serum Vitamin D Concentrations, and GeneticPolymorphisms in Vitamin D Metabolic Pathwaysin a Multicountry European StudyWilliams, K. M., Bentham, G. C. G., Young, I. S., McGinty, A., McKay, G. J., Hogg, R., Hammond, C. J.,Chakravarthy, U., Rahu, M., Seland, J., Soubrane, G., Tomazzoli, L., Topouzis, F., & Fletcher, A. E. (2016).Association Between Myopia, Ultraviolet B Radiation Exposure, Serum Vitamin D Concentrations, and GeneticPolymorphisms in Vitamin D Metabolic Pathways in a Multicountry European Study. JAMA Ophthalmology,135(1), 47-53. https://doi.org/10.1001/jamaophthalmol.2016.4752Published in:JAMA Ophthalmology

Document Version:Peer reviewed version

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

Publisher rightsCopyright 2017 American Medical Association. All rights reserved.

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:28. Aug. 2020

1

The association between myopia, ultraviolet B exposure, serum vitamin D and genetic 1

polymorphisms in vitamin D metabolic pathways in a multi country European study (EUREYE) 2

3

Katie M. Williams1,2 FRCOphth 4

Graham C.G. Bentham 3 MA 5

Ian S. Young 4 MD 6

Ann McGinty 4 PhD 7

Gareth McKay4 PhD 8

Ruth Hogg5 PhD 9

Christopher J Hammond 1 MD 10

Usha Chakravarthy 5 PhD 11

Mati Rahu 6 PhD 12

Johan Seland 7 PhD 13

Gisele Soubrane 8 MD 14

Laura Tomazzoli 9 MD 15

Fotis Topouzis10 MD 16

Astrid E. Fletcher11 PhD 17

2

18

1 Department of Ophthalmology, King's College London, UK 19

2 Department of Twin Research & Genetic Epidemiology, King's College London, UK 20

3 School of Environmental Sciences, University of East Anglia 21

4 Centre for Public Health, Queen's University Belfast, Belfast, UK 22

5 Centre for Experimental Medicine, Institute of Clinical Science, Queen's University Belfast, 23

Belfast, UK 24

6 Department of Epidemiology and Biostatistics, National Institute for Health Development, 25

Tallinn, Estonia 26

7 Eye Department, University of Bergen, Bergen, Norway 27

8 Department of Ophthalmology, Hotel Dieu de Paris, University Paris Descartes-1, France 28

9 Clinica Oculistica, Università degli Studi di Verona, Italy 29

10 Department of Ophthalmology, Aristotle University of Thessaloniki School of Medicine, 30

Greece 31

11 Faculty of Epidemiology & Population Health, London School of Hygiene and Tropical 32

Medicine, London, UK 33

34

Corresponding author 35

Professor Astrid Fletcher 36

Faculty of Epidemiology & Population Health 37

3

London School of Hygiene and Tropical Medicine 38

London 39

United Kingdom 40

Email: astrid.fletcher@lshtm.ac.uk 41

Telephone: +44 (0) 207 927 2253 42

Fax: +44 (0) 207 580 6897 43

44

Meeting Presentation 45

Parts of this material were presented at the 2016 Association for Research and Vision in 46

Ophthalmology Meeting (Abstract Number 1413) 47

48

Keywords 49

Myopia; refractive error; ultraviolet radiation; UVB; education; vitamin D; vitamin D genetics; 50

lutein 51

52

Word Count 53

2998 54

4

Revised manuscript 55

4th October 201656

5

Abstract 57

Importance: Myopia is becoming increasingly common globally and is associated with 58

potentially sight-threatening complications. Spending time outdoors is protective but the 59

mechanism underlying this association is poorly understood. 60

Objective: To examine the association of myopia with ultraviolet B (UVB) radiation (directly 61

related to time outdoors and sunlight exposure), serum vitamin D, and Vitamin D pathway 62

genetic variants, taking account of years in education. 63

Design, Setting and Participants: Cross-sectional, population-based, random sample aged ≥65 64

from six study centers from north to south Europe (EUREYE) between November 6th 2000 to 65

November 15th 2002. Of 4187 participants, 4166 attended an eye examination including 66

refraction, gave a blood sample and were interviewed by trained fieldworkers using a 67

structured questionnaire. Myopia was defined as mean spherical equivalent ≤-0.75 diopters. 68

Exclusions included aphakia, pseudophakia, late age related macular degeneration and vision 69

impairment due to cataract, resulting in 371 myopes and 2797 non-myopes (overall mean age 70

72.4 years (SD=5) and 46% male). 71

Exposures: UVB exposure estimated by combining meteorological and questionnaire data at 72

different ages, single nucleotide polymorphisms (SNPs) in Vitamin D metabolic pathway genes, 73

serum vitamin D3 (25(OH)D3), and years of education. 74

Main outcome measure: Odds Ratios (OR) of UVB, 25(OH)D3, vitamin D SNPs and myopia 75

estimated from logistic regression. 76

6

Results: A standard deviation increase in UVB exposure at ages 14-19 (0.81, 95% confidence 77

interval (CI) 0.71-0.92) and 20-39 (0.7, 95% CI 0.62-0.93) was associated with reduced adjusted 78

OR of myopia. Those in the highest tertile of years of education had twice the OR of myopia 79

(2.08, 95% CI 1.41-3.06). No independent association between myopia and 25(OH)D3 nor 80

variants in genes associated with vitamin D metabolism was found. An unexpected finding was 81

that the highest quintile of plasma lutein was associated with a reduced OR of myopia (0.57, 82

95% CI 0.46-0.72). 83

Conclusion & relevance: Increased UVB exposure was associated with reduced myopia, 84

particularly in adolescence and young adulthood. The association was not altered by 85

adjustment for education. We found no convincing evidence for a direct role of vitamin D in 86

myopia risk. The relationship between high plasma lutein and less myopia requires replication. 87

7

Myopia, or short-sightedness, is a complex trait influenced by numerous environmental and 88

genetic factors. Myopia is becoming more common worldwide, most dramatically in urban Asia 89

but prevalence rises have also been identified in the US and Europe 1,2. This has major 90

implications, both visually and financially, for the global burden from this potentially sight-91

threatening condition. 92

93

An increased risk of myopia has been associated with urbanization, higher socio-economic 94

status, prenatal factors, near work, and education 2-5. The protective effect of time outdoors on 95

myopia has been identified in studies of school age children and young adults, with replication 96

in different climates 6-10. A meta-analysis of seven cross-sectional studies concluded a 2% 97

reduced odds of myopia per additional hour of time spent outdoors per week 11. The 98

recommendation for children to spend time outdoors provides an attractive option, and 99

intervention studies are in progress 12. However, it remains unclear which of the numerous 100

elements associated with time spent outdoors, such as light intensity, ultraviolet radiation 101

(UVR) or distant focus, confers the reduced risk of myopia. Vitamin D status has been inversely 102

associated with myopia in several studies but not all 13-17, whilst genetic polymorphisms in 103

vitamin D pathway genes have been associated in one study but not in another 13,17. 104

105

We exploited the availability of relevant existing information (refractive status, UVR, education, 106

serum vitamin D and genetic polymorphisms in Vitamin D pathway genes) in the European Eye 107

Study (EUREYE) with the objective of investigating their association with myopia. 108

109

8

Methods 110

Study population 111

EUREYE was designed to maximize heterogeneity of UVR exposure and diet by selection of 112

study centers from northern to southern Europe. Participants were recruited from November 113

6th 2000 to November 15th 2002 from random sampling of the population aged over 65 years 114

in the following centers: Bergen (Norway), Tallinn (Estonia), Belfast (UK), Paris-Creteil (France), 115

Verona (Italy), Thessaloniki (Greece), and Alicante (Spain) 18. Over 11000 people were invited of 116

whom 5040 participated (45% response rate). Written informed consent was obtained from all 117

study participants. Details of study design are described elsewhere 19. Ethical approval was 118

obtained for each center from the relevant ethics committee and the research adhered to the 119

tenets of the Declaration of Helsinki. Participants attended the examination center where they 120

were interviewed by trained fieldworkers, underwent an ophthalmological examination and 121

gave a blood sample for blood measurements and genotyping. Information collected by the 122

interviewers included years of education, smoking, alcohol use, a brief medical history, a semi-123

quantitative food frequency questionnaire, and a detailed questionnaire on outdoor exposure. 124

125

Measurement of UV exposure 126

Full details of the methods have been published previously 20. Participants were sent a 127

residence and employment history to complete in advance to facilitate recall at the interview. 128

We used a questionnaire which asks about time spent outdoors between the hours of 9 a.m. 129

and 5 p.m. and between 11 a.m. and 3 p.m. daily (from the age of 14), for different 130

9

occupational and leisure time periods (including homecare), and in retirement up to current 131

age. Information from the questionnaire and residence calendar and geographical co-ordinates 132

for residence were sent to the University of East Anglia to generate estimates of individual 133

years of all day (9 a.m. to 5 p.m.) or middle of the day (11 a.m. and 3 p.m.) exposure for 134

different wavelengths of light (UVA, UVB, blue light). For all residences of one year or more, 135

ambient UVB (minimal erythema dose 21) and UVA (J cm-2) were estimated from published 136

sources that take into account time of day, month, and latitudinal variations 22. We used 137

published coefficients to adjust ambient clear-sky UV for cloud cover 23 and terrain 24. For each 138

wavelength of light, maximum potential lifetime dose was calculated as the sum of the time-139

weighted levels at each of the places of residence of the individual. Personal adult lifetime 140

(from age 14) UV exposure was estimated for each of the three wavelengths and summed for a 141

mean annual lifetime dose at different ages, for all day and middle of the day exposure. 142

143

Visual acuity and refraction 144

The protocol for testing visual acuity (VA) was different in one of the EUREYE centers (Alicante) 145

and therefore data from this center was not included in the present analysis. All other centers 146

followed the procedures described below. Presenting distance VA (i.e. with spectacles if worn) 147

was tested separately in each eye using the 4-meter EDTRS logMAR chart. Any participant who 148

was unable to achieve 0.3 logMAR (20/40 Snellen) in either eye underwent automated 149

refraction or manual retinoscopy and recording of best-corrected VA. For persons who 150

achieved 0.3 logMAR or better the spectacle correction (if any) worn by the participant for each 151

eye was measured by neutralization using a focimeter or by hand-held lenses. The spherical 152

10

equivalent (SE) was obtained by adding half of the cylindrical value to the spherical value and 153

the mean of the two eyes was calculated, commonly used in epidemiological studies. Myopia 154

was defined as SE ≤ −0.75 diopters (D) (categorized as: low SE ≤ -0.75 to > -3, moderate ≤ -3 to > 155

-6, severe ≤ -6). Non-myopes were defined as those with a SE > -0.75, and additionally those 156

with an unaided VA better than 0.3 logMAR where refraction was not measured. Participants 157

with late age-related macular degeneration (AMD), aphakia or pseudophakia in either eye or 158

visual impairment (< 0.5 logMAR or 20/60 Snellen) due to cataract were excluded. 159

160

Blood measurement 161

Blood samples were sent to a single laboratory (Queens University Belfast) for analysis. Serum 162

25-hydroxy vitamin D2 and D3 (25(OH)D2 and 25(OH)D3) were measured by liquid 163

chromatography-tandem mass spectrometry (LC-MS/MS)25. In all analyses vitamin D levels 164

were adjusted for season of measurement. Plasma lutein, zeaxanthin, beta cryptoxanthin, alpha 165

and beta-carotene, alpha and gamma tocopherol, lycopene and retinol were measured by 166

reverse phase high performance liquid chromatography. Total ascorbate was measured using 167

an enzyme-based assay in plasma stabilized with metaphosphoric acid. All assays were 168

standardized against appropriate National Institute of Standards and Technology standard 169

reference materials. Cholesterol was measured using an enzymatic assay (Randox, Crumlin, UK) 170

on a Cobas FARA centrifugal analyzer (Roche Diagnostics, UK). 171

172

Statistical analysis 173

11

Statistical analysis was carried out using Stata software version 13 (Stata Corp., College Station, 174

TX). All analyses took account of the study design of the six centers by use of robust errors. All 175

day (9 a.m. to 5 p.m.) adult lifetime UVB and 25(OH)D3 were the primary exposures of interest 176

as vitamin D3 is produced in the skin following exposure to UVB, whereas 25(OH)D2 is mainly 177

derived from fortified foods and vitamin supplements 26. Following the exclusion of 67 178

participants with very high levels, the distribution of 25(OH)D3 was normal. We investigated 179

25(OH)D3 both as a continuous variable and categorized by quintiles. Dietary vitamin D was 180

estimated using food composition tables 27 and energy adjusted. UVB was normalized using a 181

square root transformation and then Z transformed for investigation of a one standard 182

deviation (SD) increase in exposure. We calculated years of education from the difference 183

between the start and leaving dates, and categorized into tertiles in order to reflect the 184

common tiers of education (primary, secondary and higher) for inclusion as independent 185

myopia risk factor. 186

187

We ran preliminary regression analyses to identify factors associated with 25(OH)D3 and with 188

UVB as possible confounders of any association with myopia. A large number of variables were 189

independently associated with 25(OH)D3: age, sex, season, center, current smoking, diabetes, 190

obesity, dietary vitamin D intake, fish intake and fish oil supplements, antioxidants including 191

vitamin C, lutein (or zeaxanthin), retinol, alpha tocopherol, and cholesterol. Of these only lutein 192

(or zeaxanthin) were (inversely) associated with myopia and entered the models as potential 193

confounders. The factors independently associated with UVB were 25(OH)D3, center, sex and 194

12

education; only education was (positively) associated with myopia. Therefore, in our final 195

logistic regression models for myopia we retained age, sex, center and season, together with 196

our primary exposure variables (UVB, 25(OH)D3 and education) and identified confounders, 197

namely lutein (or zeaxanthin). Lutein and zeaxanthin were highly correlated (r=0.85) and results 198

were almost identical when separately introduced into the models. Our outcome measure was 199

the confounder-adjusted association between myopia and our key exposures expressed as the 200

adjusted odds ratio (OR) in logistic regression. 201

202

SNP selection, genotyping and genetic analyses 203

For reason of costs, genotyping was undertaken in a sub-sample of the main study. Data on 204

vitamin D pathway SNPs were available for a subset of 109 myopes and 782 non-myopes. 205

Ninety-three common SNPs located across 7 genes involved in the vitamin D metabolism (GC 206

(10), RXRA (14), CYP2R1 (7), DHCR7 (5), VDR (29), CYP27B1 (7), CYP24A1 (21) were selected 207

from Phase III, release 2 HapMap (http://www.hapmap.org) CEPH data (Utah residents with 208

ancestry in northern and western Europe; CEU) using Haploview 209

(http://www.broadinstitute.org/haploview) to determine linkage disequilibrium. Tag SNPs 210

were selected using multimarker tagging where r2>0.8 for all downloaded SNPs with a minor 211

allele frequency ≥5%, genotype call rate ≥95%, and no significant deviation from Hardy 212

Weinberg equilibrium (HWE). Genotyping was performed by KBiosciences (Hoddesdon, United 213

Kingdom) and associations between genotypes and myopia status were investigated. Quality 214

filters for exclusion of SNPs included call rates below 95% and deviation from HWE (P<0.001). 215

13

DNA samples were excluded if missing genotypes exceeded 10%. Other quality control 216

measures included duplicates on plates, random sample allocation to plates, independent 217

scoring of problematic genotypes by two individuals and re-sequencing of selected DNAs to 218

validate genotypes. KBiosciences quality control also included validation of all SNP assays on a 219

panel of 44 random Caucasian-derived samples plus 4 non-template (negative) controls. 220

Statistical genetic tests were performed using PLINK (version 1.07) under an additive genotypic 221

model 28. Logistic regression adjusted for age, sex, season and study center was used to 222

examine association with individual SNPs. 223

14

Results 224

The flow of participants in the study design is illustrated in Figure 1. We excluded 515 for 225

aphakia or pseudophakia, 116 for late AMD and 36 for vision impairment due to cataract. 226

Relevant exposure data (mainly serum 25(OH)D3) was missing in 297 participants (32 myopes, 227

265 non-myopes). Our final analysis was based on 371 myopes and 2797 non-myopes with 228

complete data on all relevant exposures, of which 6% (n=24) had high myopia. Included 229

participants had a mean age of 72.4 years (SD 5) and 46% were male. 230

231

In univariable analyses, there was no difference in the age or sex of people with myopia 232

compared to those without [Table 1], nor in smoking habit, alcohol use or obesity. Differences 233

were observed between myopes and non-myopes in years of education, UVB exposure and 234

serum 25(OH)D3, but there was no difference in dietary vitamin D intake. 235

236

In analyses adjusted for age, sex and center, a one SD increase in personal lifetime UVB 237

exposure was associated with reduced odds of myopia: OR = 0.72 (95% confidence interval (CI) 238

0.56-0.93), p=0.001 [Table 2]. Those in the highest tertile of years of education (median 14 239

years) had twice the odds of myopia (OR = 2.08, 95% CI 1.41-3.06, p=0.001) compared to those 240

in the lowest tertile (median 7 years). In age, sex, center and season adjusted analyses there 241

was no clear evidence for an association of 25(OH)D3 (either continuous or by quintiles) with 242

myopia. In contrast, those in the highest quintile of plasma lutein had nearly half the risk of 243

myopia (adjusted OR = 0.57, 95% CI 0.46-0.72) compared to the lowest quintile. In a further 244

15

model incorporating 25(OH)D3, lutein, education and UVB, adjusted for age, sex, center and 245

season, the estimates for each exposure were virtually unchanged. There was evidence for a 246

stronger inverse association of UVB with increasing myopia severity: low myopia OR=0.87 (95% 247

CI 0.75-1.01, p=0.06), moderate myopia OR=0.59 (95% CI 0.36-0.97, p=0.04), severe myopia 248

OR=0.39 (95% CI 0.25-0.63, p=0.001). 249

250

We investigated whether the association with myopia and UVB exposure varied by the personal 251

UVB exposure experienced at different ages. Significant ORs for less myopia with increased UVB 252

exposure were observed in adolescence and early adulthood, between the ages 14-19 and 20-253

29 [Figure 2], but not for other age groups. 254

255

The subset with genetic data were similar in age (73 years), sex (49% male) and myopia severity 256

(59% low, 34% moderate and 7% high myopia) to those without genetic data. Of the 93 genetic 257

variants associated with Vitamin D metabolism, one SNP in GC was excluded for deviation from 258

HWE. Of the remaining SNPs four were nominally associated with myopia, three in CYP2RI and 259

one in CYP24A1, but none withstood correction for multiple testing [eTable 1]. 260

261

16

Discussion 262

We found that higher annual lifetime UVB exposure, directly related to time outdoors and 263

sunlight exposure, was associated with reduced odds of myopia. UVB exposure between the 264

ages of 14-29 was associated with the highest reduction in odds of adult myopia. Myopia was 265

more than twice as common in participants in the highest tertile of education. The association 266

between UVB, education and myopia remained even after respective adjustment. This suggests 267

that the high rate of myopia associated with educational attainment is not solely mediated by 268

lack of time outdoors. 269

270

The protective effect of time outdoors on myopia is well-established 6-9,29. Time outdoors 271

reflects various physiological effects that have been associated with or hypothesized to 272

influence myopia - brighter light levels 30,31, a different spectrum of wavelengths compared to 273

artificial lighting with reduced UVR, and an extended focal distance with less hyperopic 274

peripheral defocus 32. UV conjunctival autofluorescence, an indirect marker of ocular sun 275

exposure (in particular UVR), is inversely associated with myopia 8, and has a stronger effect 276

than time outdoors assessed using questionnaires. One small study measuring UVR using 277

dosimeters found differing exposure between emmetropes, stable myopes and progressing 278

myopes 33. 279

280

Proposed mediating mechanisms include activation of dopaminergic retinal amacrine cells 281

which are stimulated by light 31 and influence ocular axial growth 34, and higher serum vitamin 282

D induced by sunlight. We, like others, did not find evidence to support the association 283

17

between myopia and serum vitamin D 16, or genes involved in vitamin D metabolism. A previous 284

publication examined 12 SNPS from two vitamin D pathway genes (VDR and GC); reporting a 285

significant association between rs2853559 in VDR in the overall sample of 289 cases and 81 286

controls and a further three variants in VDR with a low/moderate myopic subset17. In a more 287

recent publication 33 SNPs across six genes associated with vitamin D metabolism were 288

examined in over 2000 individuals in relation to both refractive error and axial length. Nominal 289

significance was identified for variants in CYP24A1 and VDR but none withstood correction for 290

multiple testing 13. We investigated the association between myopia and 92 variants in vitamin 291

D metabolism genes, identifying nominal significance in three SNPs in CYP2R1 and one SNP in 292

CYP24A1 (not the same variant as the aforementioned study). None withstood correction for 293

multiple testing. We acknowledge low power for this type of analysis but notably studied more 294

variants and previously unexamined genes (CYP2R1, RXRA) in a substantial cohort. 295

296

Those in the highest fifth of plasma lutein had approximately 40% reduced odds of myopia. 297

Results were similar for zeaxanthin but as the two are highly correlated we presented lutein 298

only for simplicity. We excluded cases of late AMD because we have previously shown an 299

increased risk for late AMD with blue light exposure in those with low levels of key antioxidants, 300

including lutein 20. Sensitivity analyses made no appreciable difference; when seventy-two 301

individuals with late AMD were included myopia OR = 0.56 (95% CI 0.46 - 0.70) with the highest 302

quintile of lutein. Lutein is a retinal carotenoid, responsible for much of the macular pigment 303

optical density (MPOD) and has anti-oxidative, anti-inflammatory and structural effects in 304

neural tissue 35. Lutein has been associated with a reduced risk of AMD 36, improved contrast 305

18

sensitivity in healthy subjects 37, and inversely related to axial length (and thus axial myopia) 38. 306

Although limited evidence for an association between lutein and myopia is gained from this 307

analysis and importantly no causative role can be inferred, it does raise interesting hypotheses 308

for a potential role. 309

310

Study Limitations 311

This study utilizes retrospectively calculated UVB exposure through highly detailed 312

questionnaires over the life course, together with geographical-specific, historical data on UVR. 313

UVB was selected for analysis given its role in stimulating vitamin D3 production in the skin. Our 314

measure is subject to recall error and lacks the heightened accuracy of UV exposure achieved 315

with light meters. However we do not have any reason to believe that the UVB association 316

would be biased since myopia was identified after the interview. A weakness of our study was 317

that we did not collect any data on UVB exposure during childhood, which could be argued to 318

be more relevant in myopia development. However, a significant proportion of refractive error 319

develops in adolescence and early adulthood 39 and our results showed the greatest effects for 320

these age groups. Non-myopes were defined either by refraction or good unaided VA where 321

refraction was unknown. This definition was used in attempt to minimize bias but to ensure this 322

was appropriate we performed sensitivity analyses where non-myopes were only classified on 323

the basis of measured refractive error; analysis using this definition produced very similar 324

results. A limitation is also that vitamin D and lutein levels were measured in later life. The 325

association between myopia development and these factors may be more relevant in younger 326

ages. However there is evidence, albeit limited, that an individual’s 25(OH)D levels are 327

19

reproducible over time 40. Variants in vitamin D pathway genes are not subject to these 328

concerns of temporality and confounding (Mendelian randomization), and hence any 329

association with myopia would strengthen a causal relationship with vitamin D. We therefore 330

consider it unlikely that vitamin D plays a role in myopia. 331

332

In conclusion this study suggests lifetime exposure of UVB is associated with reduced myopia in 333

adulthood. The protective association is strongest with exposure in adolescence and younger 334

adult life and with increasing severity of myopia. As the protective effect of time spent 335

outdoors is increasingly used in clinical interventions, a greater understanding of the 336

mechanisms and life stages at which benefit is conferred is warranted.337

20

Acknowledgements 338

EUREYE was supported by the European Commission Vth Framework (QLK6-CT-1999-02094) 339

with additional funding for cameras provided by the Macular Disease Society UK. M. Rahu was 340

financed by the Estonian Ministry of Education and Science (target funding 01921112s02 and 341

SF0940026s07). Funding for serum vitamin D analyses was provided by Guide Dogs for the Blind 342

(OR2011-05d). KMW acknowledges financial support from a Medical Research Council (UK) 343

Clinical Research Training Fellowship. No conflicting relationship exists for any author. 344

345

Financial Support 346

The sponsor or funding organization has no role in the following: design or conduct of this 347

research; collection, management, analysis, and interpretation of the data; preparation, review, 348

or approval of the manuscript; decision to submit the manuscript for publication. Full details of 349

the funding are described in the acknowledgements. 350

351

Conflicts of Interest 352

No conflicting relationship exists for any author. 353

354

Contributorship Statement 355

21

The authors’ responsibilities were as follows— KW: had the idea for the analysis and wrote the 356

paper; GB: responsible for the estimation of light exposure; IY: responsible for the antioxidant 357

assays; A McG: responsible for the vitamin D assays; GMcK: responsible for statistical genetic 358

analyses; RH: responsible for processing of refractive error data; CH; provided assistance on 359

data interpretation and drafting of the manuscript. UC, JS, MR, GS, LT, FT: responsible for the 360

acquisition of data in each of the local study centers; AEF: had full access to all of the data in 361

the study and took responsibility for the integrity of the data and the accuracy of the data 362

analysis. All authors commented on the drafts, were responsible for the decision to submit for 363

publication, and contributed to the study design. 364

365

366

22

References 367

1. Morgan IG, Ohno-Matsui K, Saw S-M. Myopia. The Lancet. 2012;379(9827):1739-1748. 368

2. Williams KM, Bertelsen G, Cumberland P, et al. Increasing Prevalence of Myopia in Europe and 369

the Impact of Education. Ophthalmology. 2015;122(7):1489-1497. 370

3. Pan CW, Ramamurthy D, Saw SM. Worldwide prevalence and risk factors for myopia. 371

Ophthalmic & physiological optics : the journal of the British College of Ophthalmic Opticians. 372

2012;32(1):3-16. 373

4. Rahi JS, Cumberland PM, Peckham CS. Myopia over the lifecourse: prevalence and early life 374

influences in the 1958 British birth cohort. Ophthalmology. 2011;118(5):797-804. 375

5. Ip JM, Saw SM, Rose KA, et al. Role of near work in myopia: findings in a sample of Australian 376

school children. Investigative ophthalmology & visual science. 2008;49(7):2903-2910. 377

6. Dirani M, Tong L, Gazzard G, et al. Outdoor activity and myopia in Singapore teenage children. 378

The British journal of ophthalmology. 2009;93(8):997-1000. 379

7. Rose KA, Morgan IG, Ip J, et al. Outdoor activity reduces the prevalence of myopia in children. 380

Ophthalmology. 2008;115(8):1279-1285. 381

8. McKnight CM, Sherwin JC, Yazar S, et al. Myopia in young adults is inversely related to an 382

objective marker of ocular sun exposure: the Western Australian Raine cohort study. American 383

journal of ophthalmology. 2014;158(5):1079-1085. 384

9. Guggenheim JA, Northstone K, McMahon G, et al. Time outdoors and physical activity as 385

predictors of incident myopia in childhood: a prospective cohort study. Investigative 386

ophthalmology & visual science. 2012;53(6):2856-2865. 387

10. Jones LA, Sinnott LT, Mutti DO, Mitchell GL, Moeschberger ML, Zadnik K. Parental history of 388

myopia, sports and outdoor activities, and future myopia. Investigative ophthalmology & visual 389

science. 2007;48(8):3524-3532. 390

23

11. Sherwin JC, Reacher MH, Keogh RH, Khawaja AP, Mackey DA, Foster PJ. The association between 391

time spent outdoors and myopia in children and adolescents: a systematic review and meta-392

analysis. Ophthalmology. 2012;119(10):2141-2151. 393

12. He M, Xiang F, Zeng Y, et al. Effect of Time Spent Outdoors at School on the Development of 394

Myopia Among Children in China: A Randomized Clinical Trial. Journal of the American Medical 395

Association. 2015;314(11):1142-1148. 396

13. Tideman JW, Polling JR, Voortman T, et al. Low serum vitamin D is associated with axial length 397

and risk of myopia in young children. European journal of epidemiology. 2016. 398

14. Choi JA, Han K, Park YM, La TY. Low serum 25-hydroxyvitamin D is associated with myopia in 399

Korean adolescents. Investigative ophthalmology & visual science. 2014;55(4):2041-2047. 400

15. Yazar S, Hewitt AW, Black LJ, et al. Myopia is associated with lower vitamin D status in young 401

adults. Investigative ophthalmology & visual science. 2014;55(7):4552-4559. 402

16. Guggenheim JA, Williams C, Northstone K, et al. Does vitamin D mediate the protective effects 403

of time outdoors on myopia? Findings from a prospective birth cohort. Investigative 404

ophthalmology & visual science. 2014;55(12):8550-8558. 405

17. Mutti DO, Cooper ME, Dragan E, et al. Vitamin D receptor (VDR) and group-specific component 406

(GC, vitamin D-binding protein) polymorphisms in myopia. Investigative ophthalmology & visual 407

science. 2011;52(6):3818-3824. 408

18. Augood CA, Vingerling JR, de Jong PT, et al. Prevalence of age-related maculopathy in older 409

Europeans: the European Eye Study (EUREYE). Archives of ophthalmology. 2006;124(4):529-535. 410

19. Augood C, Fletcher A, Bentham G, et al. Methods for a population-based study of the prevalence 411

of and risk factors for age-related maculopathy and macular degeneration in elderly European 412

populations: the EUREYE study. Ophthalmic epidemiology. 2004;11(2):117-129. 413

24

20. Fletcher AE, Bentham GC, Agnew M, et al. Sunlight exposure, antioxidants, and age-related 414

macular degeneration. Archives of ophthalmology. 2008;126(10):1396-1403. 415

21. Heckman CJ, Chandler R, Kloss JD, et al. Minimal erythema dose (MED) testing. Journal of 416

visualized experiments: JoVE. 2013(75). 417

22. Madronich S, Flocke S. Theoretical estimation of biologically effective UV radiation at the Earth’s 418

surface. In: Zerefos CS BA, ed. Solar Ultraviolet Radiation: Modeling, Measurements and Effects, 419

Vol 152. Berlin: Springer; 1997:23-48. 420

23. Frederick JE, Steele HD. The transmission of sunlight through cloudy skies: An analysis based on 421

standard meteorological information. Journal of Applied Meteorology. 1995;34(12):2755-2761. 422

24. Fesiter U, Grewe R. Spectral albedo measurements in the UV and visible region over different 423

types of surfaces. Photochemistry and Photobiology 1995;62(4):736-744. 424

25. Bennett SE, McPeake J, McCance DR, et al. Maternal vitamin D status in type 1 diabetic 425

pregnancy: impact on neonatal vitamin D status and association with maternal glycaemic 426

control. PloS one. 2013;8(9):e74068. 427

26. Reins RY, McDermott AM. Vitamin D: Implications for ocular disease and therapeutic potential. 428

Experimental Eye Research. 2015;134:101-110. 429

27. Holland B, Welch AA, Unwin ID, Buss DH, Paul AA, Southgate DAT. McCance and Widdowson's 430

The composition of Foods. Cambridge: Royal Society of Chemistry; 1991. 431

28. Purcell S, Neale B, Todd-Brown K, et al. PLINK: a tool set for whole-genome association and 432

population-based linkage analyses. American Journal of Human Genetics. 2007;81(3):559-575. 433

29. French AN, Ashby RS, Morgan IG, Rose KA. Time outdoors and the prevention of myopia. 434

Experimental Eye Research. 2013;114:58-68. 435

25

30. Smith EL, 3rd, Hung LF, Huang J. Protective effects of high ambient lighting on the development 436

of form-deprivation myopia in rhesus monkeys. Investigative ophthalmology & visual science. 437

2012;53(1):421-428. 438

31. Norton TT, Siegwart JT, Jr. Light levels, refractive development, and myopia--a speculative 439

review. Experimental Eye Research. 2013;114:48-57. 440

32. Flitcroft DI. The complex interactions of retinal, optical and environmental factors in myopia 441

aetiology. Progress in retinal and eye research. 2012;31(6):622-660. 442

33. Schmid KL, Leyden K, Chiu YH, et al. Assessment of daily light and ultraviolet exposure in young 443

adults. Optometry and vision science : official publication of the American Academy of 444

Optometry. 2013;90(2):148-155. 445

34. Stone RA, Lin T, Laties AM, Iuvone PM. Retinal dopamine and form-deprivation myopia. 446

Proceedings of the National Academy of Sciences U S A. 1989;86(2):704-706. 447

35. Johnson EJ. Role of lutein and zeaxanthin in visual and cognitive function throughout the 448

lifespan. Nutrition Reviews. 2014;72(9):605-612. 449

36. SanGiovanni JP, Chew EY, Clemons TE, et al. The relationship of dietary lipid intake and age-450

related macular degeneration in a case-control study: AREDS Report No. 20. Archives of 451

ophthalmology. 2007;125(5):671-679. 452

37. Ma L, Lin XM, Zou ZY, Xu XR, Li Y, Xu R. A 12-week lutein supplementation improves visual 453

function in Chinese people with long-term computer display light exposure. Brit J Nutr. 454

2009;102(2):186-190. 455

38. Tong N, Zhang W, Zhang Z, Gong Y, Wooten B, Wu X. Inverse relationship between macular 456

pigment optical density and axial length in Chinese subjects with myopia. Graefes Archive for 457

Clinical and Experimental Ophthalmology. 2013;251(6):1495-1500. 458

26

39. Williams KM, Hysi PG, Nag A, Yonova-Doing E, Venturini C, Hammond CJ. Age of myopia onset in 459

a British population-based twin cohort. Ophthalmic & physiological optics : the journal of the 460

British College of Ophthalmic Opticians. 2013;33(3):339-345. 461

40. Sonderman JS, Munro HM, Blot WJ, Signorello LB. Reproducibility of serum 25-hydroxyvitamin d 462

and vitamin D-binding protein levels over time in a prospective cohort study of black and white 463

adults. American Journal of Epidemiology. 2012;176(7):615-621. 464

465

27

Legends for figures 466

467

Figure 1 Flow chart of inclusion of study participants 468

469

Figure 2 Association of all-day UVB exposure at different ages with myopia, adjusted for age at 470

study examination, sex, center and years of education471

28

472

Characteristic 1

Myopia n=371

Non-myopia n=2797

P 2

Age

3 72.9 (5.5)72.4 (5.0)0.58

Men (n,%) 174 (46.9) 1282 (45.8) 0.83 Years of education

4 11 (7,14) 9 (7,12) 0.01

UVB 4,5

(Minimal Erythema Dose) 314 (140,566) 358 (224,585) 0.01 25(OH)D3 (nmol/l)

3 45.3 (20.8) 47.5 (20.9) 0.01

Dietary vitamin D mcg/day 4

1.86 (1.32, 2.62) 1.89 (1.35, 2.56) 0.62 Ever smoked (%) 179(48.2) 1350 (48.3) 0.98 Alcohol at least weekly (n,%) 134 (36.1)1106 (39.5)0.49Obesity (BMI > 30) (n, %) 138 (37.2) 1001 (35.8) 0.82 Lutein

4 (µmol/l) 0.087 (0.04, 0.24) 0.130 (0.05, 0.39) <0.01

473

Table 1 Characteristics of participants with and without myopia 474

475 1 Univariable analyses 476

2 Difference in characteristic between those with and without myopia 477

3 Mean (Standard Deviation) 478

4 Median (Inter Quartile range) 479

5 Mean annual UVB exposure 480

481

29

482

30

Characteristic Adjusted OR1

95% CI p effect 2

Adjusted OR3

95% CI p effect 2

UVB exposure (one SD increase)

0.72 0.56-0.93 0.01 0.75 0.58-0.97 0.03

Years of education (tertiles, median) 1

st, 7 1 1

2nd

, 10 1.260.99-1.580.06 1.220.96-1.570.103

rd, 14 2.08 1.41-3.06 0.001 2.04 1.40-2.96 0.001

p trend

0.001 0.0001

25(OH)D3 (continuous)

0.99 0.98-1.00 0.48

Quintiles (Q) of 25(OH)D3, median (nmol/L)

Q 1, 19.9 1 1 Q2, 33.1 0.960.79-1.310.78 0.950.74-1.220.77Q 3, 45.3 0.87 0.64-1.38 0.55 0.89 0.59-1.36 0.62 Q4, 58.9 0.750.47-1.200.24 0.780.51-1.200.28Q5, 77.0 0.87 0.51-1.47 0.60 0.87 0.56-1.38 0.59 p trend

0.31 0.31

Quintiles (Q) of plasma lutein median (µmol/L)

Q1, 0.03 1 1 Q2, 0.05 0.930.80-1.080.34 0.940.81-1.100.48Q 3, 0.11 0.82 0.55-1.20 0.30 0.83 0.55-1.25 0.39 Q4, 0.22 0.890.62-1.270.51 0.870.63-1.190.41Q5, 0.48 0.57 0.46-0.72 0.0001 0.59 0.48-0.73 <0.0001 p trend 0.0002<0.00001

31

Table 2 Association of UVB exposure, education, serum vitamin D3 (25(OH)D3) and lutein with myopia (Abbreviations: OR = odds ratio, 95% CI = 483 95% confidence interval, SD = standard deviation, Q = quintile) 484

485 1

adjusted for age, sex, center, and additionally by season for 25(OH)D3 and lutein 486 2 p effect of each variable on myopia 487

3 adjusted for age, sex, center, season, and all variables in the model (namely UVB exposure, education, 25(OH)D3, plasma lutein) 488

489

490