Myopia and the long-term incidence of cataract and cataract surgery: the Blue Mountains Eye Study

Original Article

Myopia and the long-term incidenceof cataract and cataract surgery:the Blue Mountains Eye StudyGowri L Kanthan MBBS PhD,1 Paul Mitchell MD PhD,1 Elena Rochtchina BSc MAppStat,1

Robert G Cumming MBBS PhD2 and Jie Jin Wang MMed PhD1,3

1Centre for Vision Research, Department of Ophthalmology and Westmead Millennium Institute, 2School of Public Health, University ofSydney, Sydney, and 3Centre for Eye Research, University of Melbourne, Melbourne, Australia

ABSTRACT

Background: To assess the association between refrac-tive errors and the 10-year incidence of cataract andcataract surgery.

Design: Population-based prospective study.

Participants: Three thousand six hundred fifty-fourpersons aged 49+ years living in a well-defined geo-graphical region were examined at baseline; 2564were re-examined after 5 and/or 10 years.

Methods: Baseline refractive error was measuredusing autorefraction with subjective refinement.Lens photographs were taken at each visit andassessed using the Wisconsin Cataract GradingSystem.

Main Outcome Measures: Long-term incidence ofcataract and cataract surgery.

Results: Compared with emmetropia, high myopiawas associated with increased incidence of nuclearcataract (adjusted odds ratio 3.01, 95% confidenceintervals 1.35–6.71). Low (odds ratio 1.86, confi-dence interval 1.03–3.35) and high myopia (oddsratio 7.80, confidence interval 3.51–17.35) were sig-nificantly associated with higher incidence of pos-terior subcapsular cataract. Low, moderate and high

myopia were associated with increased incidence ofcataract surgery (odds ratio 2.54, confidence interval1.76–3.68; odds ratio 2.61, confidence interval 1.45–4.69; and odds ratio 4.81, confidence interval 2.33–9.93, respectively). Either any (odds ratio 1.35,confidence interval 1.08–1.69) or moderate hypero-pia (odds ratio 1.76, confidence interval 1.32–2.34)was associated with increased incidence of nuclearcataract.

Conclusion: Our longitudinal study confirms theassociation between myopia and an increased risk ofnuclear and posterior subcapsular cataract. It alsosuggests that hyperopia may increase the risk ofnuclear cataract.

Key words: Blue Mountains Eye Study, cataract, cata-ract surgery, refractive error.

INTRODUCTION

Refractive error is frequent in older people affectingapproximately two thirds of the population.1 Cata-ract is another significant ocular disease affecting theaging population.2 Improved understanding of riskfactors for cataract would help to identify high-riskgroups and assist in eye health-care planning.

A number of cross-sectional studies have sug-gested an association between refractive error and

■ Correspondence: Dr Jie Jin Wang, Center for Vision Research, Westmead Hospital, Hawkesbury Road, Westmead, Sydney, 2145, Australia. Email:

[email protected]

Received 5 May 2013; accepted 12 August 2013.

Competing/conflicts of interest: No stated conflict of interest.

Funding sources: The study was supported by the Australian National Health & Medical Research Council, Canberra, Australia (Grant nos.

974159 and 211069).

bs_bs_banner

Clinical and Experimental Ophthalmology 2014; 42: 347–353 doi: 10.1111/ceo.12206

© 2013 Royal Australian and New Zealand College of Ophthalmologists

mailto:[email protected]

cataract.3–9 Findings from the two short-term longi-tudinal studies to date have been inconsistent.10,11

We therefore aimed to examine the relationshipbetween refractive error at baseline and the long-term (10-year) incidence of cataract subtypes andcataract surgery in the Blue Mountains Eye Study(BMES) population.

METHODS

Study population

Details of the BMES population and its methodsare reported elsewhere.2 In brief, the BMES is apopulation-based cohort study of vision andcommon eye diseases in an urban older populationcomprising two postcode areas in the Blue Moun-tains region, west of Sydney, Australia. This geo-graphically well-defined area has a stable populationthat is reasonably representative of Australia insocioeconomic status and other measures. All resi-dents of these two postcode areas who were aged 49years or older were eligible and were invited to par-ticipate in the survey.

At baseline examinations (1992–1994), 4433residents were identified as eligible to participate,of whom 3654 (82.4%) were interviewed andexamined. Differences between participants andnon-participants at baseline have previously beenreported.12 All surviving participants were invitedfor re-examination after 5 (1997–1999) and 10years (2002–2004), with 2335 (75.1% of survivors)and 1952 (75.6% of survivors) returning for re-examinations at these times, respectively. Altogether,2564 participants were re-examined at least oncesince the baseline examination. The study adheredto recommendations of the Helsinki Declarationand was approved by the University of Sydneyand Western Sydney Area Health Service HumanResearch Ethics Committees. Written informedconsent was obtained from all participants.

Procedures

An interviewer-administered questionnaire wasused to collect detailed demographic and medicalhistory at each visit. Objective refraction was per-formed with an autorefractor (model 530; Humphrey,San Leandro, CA, USA). This was followed by sub-jective refraction according to the Beaver Dam EyeStudy (BDES) modification of the Early TreatmentDiabetic Retinopathy Study protocol using a loga-rithm of minimum angle of resolution chart.

All participants underwent detailed eye exami-nations. Slit-lamp lens photographs were taken ofeach eye on Ektachrome 200 colour film (Kodak,Rochester, NY, USA) and a Topcon SL-7E photograph

slit-lamp camera (Topcon, Tokyo, Japan) to assesspresence of nuclear cataract. Retro-illumination lensphotographs were taken using a Neitz CT-R cataractcamera (Neitz Instruments, Tokyo, Japan) to assesspresence of cortical and posterior subcapsular (PSC)cataract. The Wisconsin Cataract Grading System,13

first developed in 1990 for use in the BDES, wasclosely followed in masked grading of all lens photo-graphs. Details of this grading method are describedelsewhere.13 Intergrader and intragrader reproduc-ibility of the lens grading was assessed using quad-ratic weighted kappa statistics and was shown to bewithin an acceptable range in our study.2 At the 5-and 10-year follow-up visits, participants werere-examined approximately in the same order as atbaseline using the same procedures and equipment.

Definitions

The baseline refraction state was defined as thespherical equivalent of refraction, calculated by thealgebraic addition of the best-corrected sphericalrefraction and half the cylindrical refraction.Emmetropia was defined as a spherical equivalent ofrefraction between +1 Dioptre (D) and −1 D inclu-sive, hyperopia as more than +1 D and myopia asless than −1 D. Myopia was further classified as low(less than −1 D to more than −3.5 D), moderate(−3.5 D or less to more than −6 D) and high (−6 D orless). Hyperopia was further classified as low(greater than +1 D and less than +2 D), moderate(+2 D to less than +4 D) and high (+4 D and higher).

Incident cataract was defined as the appearance ofnuclear, cortical or PSC cataract subtypes in phakiceyes, in which the corresponding cataract subtypewas not present at baseline. Similarly, incident cata-ract surgery was defined as cataract surgery per-formed during the follow-up period in eyes thatwere phakic at baseline.

Statistical analysis

SAS software (SAS Institute, Cary, NC, USA) wasused for data analysis. Eye-specific data and gener-alized estimating equation models were used to esti-mate the associations between baseline refractiveerror and the 10-year incidence of cortical, nuclear,and PSC cataract or cataract surgery, while control-ling for correlation between two eyes. Odds ratios(ORs) and 95% confidence intervals (CIs) are pre-sented for the associations. Adjustments were madefor potential confounders, namely age, gender,smoking, diabetes, socioeconomic status and cortico-steroid use, in multivariable-adjusted models.

RESULTS

Table 1 shows the associations between refractiveerror and long-term incidence of nuclear cataract.

348 Kanthan et al.


Compared with emmetropia, high myopia was asso-ciated with increased incidence of nuclear cataractafter multivariable adjustment (OR 3.01, 95% CI1.35–6.71). Any hyperopia and moderate hyperopiawere also associated with increased incidence ofnuclear cataract after adjustment in a multivariablemodel (OR 1.35, 95% CI 1.08–1.69; and OR 1.76,95% CI 1.32–2.34, respectively).

Table 2 shows the associations between refractiveerror and the long-term incidence of cortical cataract.Eyes with moderate myopia at baseline had a signifi-cantly higher incidence of cortical cataract comparedwith emmetropic eyes after adjusting for potentialconfounders (OR 1.79, 95% CI 1.04–3.08). This asso-ciation was not observed for eyes with either low orhigh myopia, assessed separately. No associations

were found between hyperopia and incident corticalcataract.

Table 3 shows the associations between refractiveerror and long-term incidence of PSC cataract. Eyeswith any myopia, low myopia and high myopia hada significantly higher incidence of PSC cataract com-pared with emmetropic eyes (OR 2.12, 95% CI 1.32–3.40; OR 1.86, 95% CI 1.03–3.35; and OR 7.80, 95%CI 3.51–17.35, respectively) after multivariableadjustment. Eyes with moderate hyperopia had asignificantly lower incidence of PSC cataract com-pared with emmetropic eyes after adjusting forage and gender (OR 0.59, 95% CI 0.39–0.99). Thisassociation, however, was no longer statistically sig-nificant after adjustment for additional confounders(OR 0.59, 95% CI 0.34–1.04).

Table 1. Associations (odds ratios [ORs] and 95% confidence intervals [CIs]) between baseline refraction and incident nuclear cataract

Refractive state Incident rates Age-sex adjusted Multivariate-adjusted†

No of eyes at risk Percentage withincident cataract

OR 95% CI OR 95% CI

Emmetropia 1216 18.0 1.0 Referent 1.0 ReferentHyperopia§

Any 1233 32.9 1.36 (1.10–1.69) 1.35 (1.08–1.69)Low 696 26.7 1.15 (0.90–1.46) 1.15 (0.89–1.48)Moderate 471 42.7 1.82 (1.38–2.39) 1.76 (1.32–2.34)High 66 28.8 1.17 (0.66–2.06) 1.19 (0.67–2.10)

Myopia‡


†Adjusted for age, sex, smoking, diabetes, socioeconomic status and steroid use. ‡Low myopia, less than −1 to more than −3.5 D;moderate myopia, −3.5 to more than −6 D; high myopia, −6D or less. §Low hyperopia, greater than +1 D to less than +2 D; moderatehyperopia, +2 D to less than +4 D; high hyperopia, +4 D or greater. Significant ORs (P > 0.05) are bolded.

Table 2. Associations (odds ratios [ORs] and 95% confidence intervals [CIs]) between baseline refraction and incident cortical cataract



OR 95% CI OR 95% CI



Myopia‡


†Adjusted for age, sex, smoking, diabetes, socioeconomic status and steroid use. ‡Low myopia, less than −1 to more than −3.5 D;moderate myopia, −3.5 to more than −6 D; high myopia, −6 D or less. §Low hyperopia, greater than +1 D to less than +2 D; moderatehyperopia, +2 D to less than +4 D; high hyperopia, +4 D or greater. Significant ORs (P > 0.05) are bolded.

Myopia and incidence of cataract 349


Table 4 shows the associations between refractiveerror at baseline and the long-term incidence of cata-ract surgery. In a multivariable model, all myopialevels (low, moderate and high) were significantlyassociated with incident cataract surgery. The riskprogressively increased with increasing degree ofmyopia (OR 2.54, 95% CI 1.76–3.68 for low myopia;OR 2.61, 95% CI 1.45–4.69 for moderate myopia;and OR 4.81, 95% CI 2.33–9.93 for high myopia). Noassociations were found between hyperopia and theincidence of cataract surgery.

Table 5 shows the associations between refractiveerror as a continuous variable and the long-term inci-dence of cataract or cataract surgery. After adjustmentfor potential confounders, hyperopia was signifi-cantly associated with the long-term incidence of

nuclear cataract (OR 1.15, 95% CI 1.06–1.24 perdioptre) compared with emmetropia. Myopia wassignificantly associated with the long-term incidenceof nuclear cataract (OR 1.14, 95% CI 1.02–1.26 perdioptre), PSC cataract (OR 1.20, 95% CI 1.08–1.33per dioptre) and cataract surgery (CI 1.19, 95% CI1.10–1.29 per dioptre) compared with emmetropia.

DISCUSSION

In this cohort study of an older Australians, myopiaat baseline was found to be associated with the long-term (10-year) incidence of nuclear cataract, PSCcataract and cataract surgery. Hyperopia at baselinewas found to be associated with an increased risk ofnuclear cataract.

Table 3. Associations (odds ratios [ORs] and 95% confidence intervals [CIs]) between baseline refraction and incident posteriorsubcapsular cataract



OR 95% CI OR 95% CI



Myopia‡



Table 4. Associations (odds ratios [ORs] and 95% confidence intervals [CIs]) between baseline refraction and incident cataract surgery


No of eyes at risk Percentage with incidentcataract surgery

OR 95% CI OR 95% CI

Emmetropia 2122 9.2 1.0 referent 1.0 ReferentHyperopia§


Myopia‡



350 Kanthan et al.


Several population based cross-sectional studiesincluding the Melbourne Visual ImpairmentProject,5,14 Barbados Eye Study,7 BDES10 and BMES15

have shown an association between myopic refrac-tion and nuclear cataract. This association has beenhypothesized to reflect increasing refractive indexof the lens from nuclear sclerosis causing a myopicshift in refraction (index myopia) in older agedpersons. The hypothesis was confirmed in BDES lon-gitudinal data that showed that eyes with nuclearcataract at baseline had a myopic shift in refraction atlong-term follow up, whereas eyes without signi-ficant nuclear cataract at baseline had a hyperopicshift in refraction at follow up.16 BMES long-termfollow-up data also confirmed such a temporal asso-ciation between baseline nuclear sclerosis and sub-sequent myopic shift.17 Our current study showed anassociation in the opposite direction, as eyes withhigh myopia at baseline had an increased risk ofdeveloping nuclear cataract over the longer term.Although this finding was not reported from theBDES population,10 it raises the possibility thatwhereas nuclear cataract causes a myopic shift inrefraction, high myopia likely increases the risk ofdeveloping nuclear cataract.

BDES short-term follow-up data showed thathyperopia was also associated with an increasedincidence of nuclear cataract.10 This association,although not observed in our baseline (cross-sectional) or 5-year (short-term) follow-up analyses,is supported by our current long-term follow-updata. The odds of developing nuclear cataract were35% higher in eyes with any hyperopia comparedwith eyes with emmetropia.

BDES longitudinal data also showed a significanttrend between increasing degrees of hyperopia andthe incidence of cortical cataract. Our current study,however, failed to demonstrate any associationbetween hyperopia and cortical cataract. PreviouslyBMES cross-sectional data showed that high myopiawas associated with cortical cataract.15 Short-termfollow-up data from the BMES11 and BDES,10

however, did not show any association between

myopia and the 5-year incidence of cortical cataract.Similarly, our current long-term follow-up datafailed to demonstrate any consistent associationsbetween myopia and cortical cataract.

Cross-sectional BMES data previously showedthat hyperopic eyes had reduced risk of PSC cata-ract.15 Our current longitudinal data, however, didnot confirm this association. Our previous reportsshowed that myopic eyes had significantly higherprevalence as well as 5-year incidence of PSC cata-ract.11,15 Our current long-term follow-up dataconfirm this strong association between myopia andincident PSC cataract. Although the BDES data didnot show a significant association between myopiaand the prevalence or short-term incidence of PSCcataract,10 it did provide evidence for a significantassociation between myopia and the incidence ofcataract surgery.10 The lack of association foundbetween myopia and the incidence of PSC cataract inthe BDES could also have been caused by the effectsof cataract surgery during follow up. Because of itscentral location and rapid progression, PSC cataractis the most visually disabling cataract subtype andfrequently leads to early cataract surgery.2

Various biological mechanisms may explain theassociation between cataract and refractive error.The lens normally exists in a low oxygen environ-ment18 and increased exposure of the lens to oxygenappears to cause cataract.19,20 The vitreous gelensures that oxygen from the retina can only reachthe lens by slow diffusion. Replacement of the vit-reous gel with liquid, as occurs with vitreoussyneresis will increase the fluid circulation withinthe vitreous chamber and readily distribute oxygenfrom the retinal surface to the lens. It has beenshown that the extent of vitreous liquefaction waspositively correlated with the extent of nuclearopacification even after adjusting for age.21 In myo-pic eyes, vitreous syneresis occurs at an earlier age,is more extensive and increases with the degree ofmyopia.22 As a result, myopic eyes are vulnerable tooxidative damage and subsequent nuclear cataractformation.

Table 5. Associations (odds ratios [ORs] and 95% confidence intervals [CIs], per dioptre deviation from emmetropia) between baselinerefraction as a continuous variable and the incidence of cataract

Refractive Incident cataract

Deviation towards Nuclear Cortical PSC Cataract surgery

HyperopicAge- and gender- adjusted OR (CI) 1.16 (1.07–1.25) 1.01 (0.93–1.10) 0.85 (0.69–1.06) 1.02 (0.92–1.13)Multivariable-adjusted OR (CI)† 1.15 (1.06–1.24) 1.01 (0.93–1.11) 0.82 (0.65–1.03) 1.04 (0.94–1.16)

MyopicAge- and gender- adjusted OR (CI) 1.10 (1.00–1.20) 1.04 (0.97–1.12) 1.22 (1.11–1.33) 1.18 (1.10–1.26)Multivariable-adjusted OR (CI)† 1.14 (1.02–1.26) 1.05 (0.97–1.14) 1.20 (1.08–1.33) 1.19 (1.10–1.29)

†Adjusted for age, sex, smoking, diabetes, socioeconomic status and steroid use. Significant ORs (P > 0.05) are bolded. PSC, posteriorsubcapsular.



Glutathione is an anti-oxidant found in the eyethat protects the lens from oxidative damage.23 Com-pared with healthy control subjects, cataractouslenses were found to have lower levels of gluta-thione, with the lowest levels found in myopiclenses.24 Melondialdehyde is a breakdown productof lipid peroxidation, and its presence in the vitreoussuggests a retinal source. Rod outer segments areparticularly susceptible to lipid peroxidationbecause of the high concentration of polyunsaturatedlipid in their membranes. Increased levels ofmelondialdehyde have been found in cataractouslenses and in the vitreous of myopes compared withcontrol subjects and non-myopic cataract lenses.24,25

In addition, a positive correlation has been foundbetween the level of thiobarbituric acid reactive sub-stances (also indicating lipid peroxidation) in thesubretinal fluid of patients undergoing retinaldetachment surgery and the degree of myopia.26

These findings suggest that increasing myopia maybe related to increasing damage to rod outer seg-ments; the by-products of this process may then con-tribute to cataract formation.

Strengths of our study include its long-termfollow up of participants with reasonable rates ofsurveillance, consistency of the methods used atbaseline and follow-up examinations to assess cata-ract, masked grading of lens photographs and docu-mentation of the many potential confoundersadjusted for in our analyses. Limitations of ourstudy should also be noted. The lack of clear dose–response relationship in some of the observed asso-ciations raises a question whether these associationsare real, and consistency in observing the sameassociations across different study populations infuture will help to enhance confidence in these find-ings. Given the large number of associations exam-ined in this study, some of the positive associationscould have been purely due to chance. Neverthe-less, even after Bonferroni correction for multipletesting, the association of nuclear cataract withhyperopia and the associations of PSC cataract andcataract surgery with myopia remained significant(corrected P values 0.008, 0.007 and 0.001, respec-tively). On the other hand, some negative associa-tions could have occurred due to lack of studypower. This is particularly applicable to PSC cata-ract, which has a relatively low long-term incidencein our study population,2 likely as a result of cata-ract surgery during the interval before follow-upvisits.

In summary, long-term follow-up data from thispopulation-based, older Australian cohort provideevidence to support the hypothesis that myopia con-tributes to an increased risk of nuclear and PSC cata-ract subtypes, whereas hyperopia increases the riskof nuclear cataract.

REFERENCES

1. Attebo K, Ivers RQ, Mitchell P. Refractive errors in anolder population: the Blue Mountains Eye Study. Oph-thalmology 1999; 106: 1066–72.

2. Kanthan GL, Wang JJ, Rochtchina E et al. Ten-yearincidence of age-related cataract and cataract surgery inan older australian population: the Blue MountainsEye Study. Ophthalmology 2008; 115: 808–14.

3. Giuffr G, Dardanoni G, Lodato G. A case-control studyon risk factors for nuclear, cortical and posteriorsubcapsular cataract: the Casteldaccia Eye Study. ActaOphthalmol Scand 2005; 83: 567–73.

4. Harding JJ, Harding RS, Egerton M. Risk factors forcataract in Oxfordshire: diabetes, peripheral neuropa-thy, myopia, glaucoma and diarrhoea. Acta Ophthalmol(Copenh) 1989; 67: 510–7.

5. Wensor M, McCarty CA, Taylor HR. Prevalence andrisk factors of myopia in Victoria, Australia. ArchOphthalmol 1999; 117: 658–63.

6. Wong TY, Foster PJ, Hee J et al. Prevalence andrisk factors for refractive errors in adult Chinese inSingapore. Invest Ophthalmol Vis Sci 2000; 41: 2486–94.

7. Wu SY, Nemesure B, Leske MC. Refractive errors in ablack adult population: the Barbados Eye Study. InvestOphthalmol Vis Sci 1999; 40: 2179–84.

8. Dandona R, Dandona L, Naduvilath TJ, Srinivas M,McCarty CA, Rao GN. Refractive errors in an urbanpopulation in Southern India: the Andhra Pradesh EyeDisease Study. Invest Ophthalmol Vis Sci 1999; 40:2810–8.

9. Chang MA, Congdon NG, Bykhovskaya I et al. Theassociation between myopia and various subtypes oflens opacity: SEE (Salisbury Eye Evaluation) project.Ophthalmology 2005; 112: 1395–401.

10. Wong TY, Klein BE, Klein R, Tomany SC, Lee KERefractive errors and incident cataracts: the BeaverDam Eye Study. Invest Ophthalmol Vis Sci 2001; 42: 1449–54.

11. Younan C, Mitchell P, Cumming RG, Rochtchina E,Wang JJ. Myopia and incident cataract and cataractsurgery: the Blue Mountains Eye Study. InvestOphthalmol Vis Sci 2002; 43: 3625–32.

12. Mitchell P, Cumming RG, Attebo K, Panchapakesan J.Prevalence of cataract in Australia. Ophthalmology 1997;104: 581–8.

13. Klein BEK, Klein R, Linton KLP, Magli YL, NeiderMW. Assessment of cataracts from photographs in theBeaver Dam Eye Study. Ophthalmology 1990; 97: 1428–33.

14. McCarty CA, Mukesh BN, Fu CL, Taylor HR. The epi-demiology of cataract in Australia. Am J Ophthalmol1999; 128: 446–65.

15. Lim R, Mitchell P, Cumming RG. Refractive associa-tions with cataract: the Blue Mountains Eye Study.Invest Ophthalmol Vis Sci 1999; 40: 3021–6.

16. Lee KE, Klein BE, Klein R et al. Changes in refractionover 10 years in an adult population: the Beaver DamEye study. Invest Ophthalmol Vis Sci 2002; 43: 2566–71.

352 Kanthan et al.


17. Fotedar R, Mitchell P, Burlutsky G, Wang JJ. Relation-ship of 10-year change in refraction to nuclear cataractand axial length findings from an older population.Ophthalmology 2008; 115: 1273–8.

18. Holekamp NM, Shui YB, Beebe DC. Vitrectomysurgery increases oxygen exposure to the lens: a pos-sible mechanism for nuclear cataract formation. Am JOphthalmol 2005; 139: 302–10.

19. Palmquist BM, Philipson B, Barr PO. Nuclear cataractand myopia during hyperbaric oxygen therapy. Br JOphthalmol 1984; 68: 113–7.

20. Boscia F, Grattagliano I, Vendemiale G, Micelli-FerrariT, Altomare E. Protein oxidation and lens opacity inhumans. Invest Ophthalmol Vis Sci 2000; 41: 2461–5.

21. Harocopos GJ, Shui YB, McKinnon M, Holekamp NM,Gordon MO, Beebe DC. Importance of vitreous lique-faction in age-related cataract. Invest Ophthalmol Vis Sci2004; 45: 77–85.

22. Soubrane G, Coscas. G, Kuhn D Myopia. In: GuyerDR, Yannuzzi L, Chang S, Shields JA, Green WR, eds.Myopia. Philadelphia, PA: WB Saunders Co, 1999; 189–205.

23. Kamei A. Glutathione levels of the human crystallinelens in aging and its antioxidant effect against theoxidation of lens proteins. Biol Pharm Bull 1993; 16:870–5.

24. Micelli-Ferrari T, Vendemiale G, Grattagliano I et al.Role of lipid peroxidation in the pathogenesis ofmyopic and senile cataract. Br J Ophthalmol 1996; 80:840–3.

25. Simonelli F, Nesti A, Pensa M et al. Lipid peroxidationand human cataractogenesis in diabetes and severemyopia. Exp Eye Res 1989; 49: 181–7.

26. Bosch-Morell F, Sanz A, Diaz-Llopis M, Romero FJ.Lipid peroxidation products in human subretinalfluid. Free Radic Biol Med 1996; 20: 899–903.



Documents

Myopia and the long-term incidence of cataract and cataract surgery: the Blue Mountains Eye Study