The Relationship Between Glaucoma and Myopia the Blue Mountains Eye Study

The Relationship between Glaucoma and Myopia The Blue Mountains Eye Study

Paul Mitchell, MD, FRCOphth, Fleur Hourihan, BSc, MPH, Jen Sandbach, MB, FRACO, Jie Jin Wang, MB, MMed (ClinEpi)

Objective: To quantify the relationship between myopia and open-angle glaucoma, ocular hypertension (OH), and intraocular pressure (IOP) in a representative older population. Design: Cross-sectional population-based study of 3654 Australians 49 to 97 years of age. Methods: Subjects with any myopia (Ն-1.0 diopter [D]) were identified by a standardized subjective refraction and categorized into low myopia (Ն-1.0 D to Ͻ-3.0 D) or moderate-to-high myopia (Ն-3.0 D). Glaucoma was diagnosed from characteristic visual field loss, combined with optic disc cupping and rim thinning, without reference to IOP. Ocular hypertension was diagnosed when applanation IOP was greater than 21 mmHg in either eye in the absence of glaucomatous visual field and optic disc changes. Main Outcome Measure: General estimating equation models were used to assess associations between eyes with myopia and either glaucoma or OH. Results: Glaucoma was present in 4.2% of eyes with low myopia and 4.4% of eyes with moderate-to-high myopia compared to 1.5% of eyes without myopia. The relationship between glaucoma and myopia was maintained after adjusting for known glaucoma risk factors, odds ratio (OR) of 2.3, and 95% confidence intervals (CI) of 1.3 to 4.1 for low myopia. It was stronger for eyes with moderate-to-high myopia (OR, 3.3; CI, 1.7–6.4). Only a borderline relationship was found with OH, OR of 1.8 (CI, 1.2–2.9) for low myopia, and OR of 0.9 (CI, 0.4–2.0) for moderate-to-high myopia. Mean IOP was approximately 0.5 mmHg higher in myopic eyes compared to nonmyopic eyes. Conclusions: This study has confirmed a strong relationship between myopia and glaucoma. Myopic subjects had a twofold to threefold increased risk of glaucoma compared with that of nonmyopic subjects. The risk was independent of other glaucoma risk factors and IOP. Ophthalmology 1999;106:2010–2015

An association between myopia and primary open-angle ofNorthAfricanorAsianorigin.17Otherstudieshave glaucomahasbeenrecognizedfordecades1–3anddocu- reportedhigherapplanationpressuresinmyopicsubjects,15 mentedinnumerouscaseseries4–9andinmost,10,11butnot includingchildren,19orinsubjectswithincreasedaxial all,12case–controlstudies.Otherreportshavehighlighted length.14However,norelationshipwithIOPwasfoundina thehighfrequencyofmyopiainyoungadultspresenting UnitedKingdomstudy10orinmyopicanisometropia,21 13 withopen-angleglaucoma. Somestudieshavefoundthe whereas one study reported an association between myopia 9 relationshiponlyinpatientswithhighmyopia. andlow-tensionglaucoma.22 Myopia has also been found to have an influence on 14–20 Selection bias could account for some of the reported intraocularpressure(IOP). Inalargecase–control association between glaucoma and myopia in case series study, myopic refractive error was found to be strongly 16 and case–control studies, as myopic subjects are likely to associatedwithocularhypertension(OH). AnIsraeli seek ophthalmic care more frequently and glaucoma is a study of 2403 subjects documented a significant relationship relativelyunderdiagnosedconditioninthecommunity.23,24 between myopia and increasing IOP, particularly in persons This bias may be lessened in population studies, in which the diagnosis of disease is based on a masked assessment of diagnostic characteristics. Originally received: October 29, 1997. To date, however, no large population-based studies Revision accepted: June 11, 1999. Manuscript no. 97755. have examined the role and relative strength of myopia as a From the Department of Ophthalmology, University of Sydney, Westmead risk factor for glaucoma. The “use of eyeglasses for read- Hospital, Sydney, Australia. ing” was assessed in the Barbados Eye Study, but the Supported by the Australian Department of Health and Family Services relationship between open-angle glaucoma and myopia was and the Save Sight Institute, University of Sydney, Sydney, Australia. notassessed.25Wethereforeaimedtoinvestigateassocia- Reprint requests to Paul Mitchell, MD, FRCOphth, University of Sydney Department of Ophthalmology, Hawkesbury Rd, Westmead, NSW, 2145, tions between myopia and open-angle glaucoma, OH, and Australia.E-mail:[email protected]. IOP in a well-defined older population, in which both re-

2010 Mitchell et al ⅐ Relationship between Glaucoma and Myopia fractiveerror26andglaucomaprevalence24wereassessed sion was defined as a history of hypertension with current use of and other glaucoma risk factor data were collected. antihypertensive medication or elevated blood pressure (systolic Ն 160 mmHg or diastolic Ն 95 mmHg). The questionnaire assessed familyhistoryofglaucoma,historyoftypicalmigraine,28and historyofinhaledorsystemicsteroiduse.29Pseudoexfoliationwas Methods assessed from the slit-lamp examination. Statistical Analysis System (SAS; SAS Institute, Cary, NC) was The Blue Mountains Eye Study is a population-based survey of used for statistical analyses, including Mantel–Haenszel chi-square age-related eye diseases in residents of an urban population in the test for trend and generalized linear models. Associations between Blue Mountains region, west of Sydney, Australia. Survey meth- glaucoma and myopia were assessed for individual eyes, because odsweredescribedpreviously.24,26,27Allpermanentnoninstitu- of the asymmetry of refractive error, glaucomatous damage, or tionalizedresidents49yearsofageorolderwereidentifiedina IOP in some subjects. The Liang and Zeger general estimating door-to-door census. Of 4433 eligible residents, 3654 (82.4%) equationmethod,30whichtakesintoaccountthecorrelationbe- were examined from 1992 to 1994. After excluding persons who tween eyes, was used to assess relationships between these vari- died or left the area during the survey and could not be examined, ables. Analyses by subject were also performed using logistic the response rate was 87.9%. The study was approved by the regression. Age and IOP were used as continuous variables. Odds Western Sydney Area Human Ethics Committee, and written, ratios (OR) and 95% confidence intervals (CI) are presented. informed consent was obtained from all participants. An interviewer-administered questionnaire included demo- graphic characteristics, medication use, visual function, and med- ical history including diabetes, hypertension, and migraine. Sub- Results jects underwent a detailed eye examination, which included applanation tonometry, Humphrey automated perimetry, stereo- Definite or probable glaucoma was diagnosed in 108 persons (3%); scopicopticdiscphotography,slit-lampexamination,andsubjec- theprevalenceincreasedexponentiallywithage.24Amongpersons tiverefraction.24,26TheHumphrey76-pointsuprathresholdvisual diagnosed with glaucoma, 72 (66.7%) had typical glaucomatous field test (Allergan Humphrey, San Leandro, CA) was performed field loss in the right eye and 88 (81.5%) had field loss in the left in 3241 (89%) participants, of whom 352 persons (9.6%) were eye (total, 160 eyes). After nonphakic eyes were excluded, a total classified as glaucoma suspects, either from defects on the 76-point of 126 eyes had diagnostic glaucomatous field loss and could be test or because optic disc signs suggested glaucoma. All were used for analysis. The OH was found in 135 participants, a prev- asked to return for Humphrey full-threshold 30–2 visual fields, of alence of 3.7% (CI, 3.1–4.3) but with no significant age-related whom 336 (9.2% of the population) completed the test. Stereo increase in prevalence.24 Among persons with OH, 124 (91.9%) optic disc photographs of both eyes (30°) were performed in 98% had elevated IOP in the right eye and 109 (80.7%) had elevated of subjects. IOP in the left eye (total, 233 eyes). After excluding nonphakic Open-angle glaucoma (here termed glaucoma) was diagnosed eyes, a total of 211 eyes had OH and were available for analysis. by the presence of matching optic disc cupping with rim thinning After nonphakic eyes were excluded, 866 eyes (12.5%) with (cup–disc ratio Ն 0.7 or cup–disc asymmetry Ն 0.3) and charac- anymyopia(SEqՆ-1.0D)wereavailableforanalysis(Table1). teristic visual field loss on automated perimetry, after excluding Low myopia (Ն-1.0 D to Ͻ-3.0 D) was present in 524 eyes rubeotic, secondary, or angle-closure glaucoma with gonioscopy. (7.6%), and moderate-to-high myopia (SeEq Ն-3.00 D) was found ThediagnosisofglaucomawasmadewithoutreferencetoIOP.24 in 342 eyes (4.9%). A higher age-adjusted prevalence of myopia Characteristicglaucomatousfieldlosswasdefinedasanabnormal was found in women than in men (OR, 1.4; CI, 1.0–2.0), and the 2 Humphrey 30–2 Glaucoma Hemifield Test with one or more of the prevalence decreased significantly with increasing age (0 trend ϭ following defects not explained by ocular or neurologic causes: (1) 10.73, 1 degree of freedom, P Ͻ 0.001). arcuate or paracentral scotoma, at least four contiguous points on A strong association was found between low myopia and the pattern deviation plot depressed at P Ͻ 0.5% level; (2) nasal glaucoma,asshowninTable2.Glaucomatousdamagetotheoptic step at least two horizontal points in width (10°) on the pattern disc and visual field was more than twice as frequent in eyes with deviation plot depressed at P Ͻ 0.5% level; or (3) advanced low myopia (4.2%) than in eyes without myopia (1.5%) (OR, 2.1; glaucomatous field loss. The OH was defined as an IOP greater CI, 1.2–3.8) after adjusting for age and gender. This relationship than 21 mmHg in either eye, but without diagnostic visual field and remained after simultaneously adjusting for other known glaucoma optic disc signs, after excluding persons with open-angle or other risk factors as well (e.g., glaucoma family history, diabetes, hy- forms of glaucoma. pertension, history of typical migraine, steroid use, and presence of A Humphrey autorefractor (Model 530) was used to obtain an pseudoexfoliation) (OR, 2.3; CI, 1.3–4.1). It was also consistently objective refraction. Subjective refraction was then performed present across the age groups examined, apart from the youngest using the Beaver Dam Eye Study modification of the Early Treat- age group, which had inadequate statistical power because of small ment Diabetic Retinopathy Study protocol with a logarithm of the numbers. minimumangleofresolution(LogMAR)chart.26Sphericalequiv- A stronger association was found between moderate-to-high alent refractive error was calculated as (sphere ϩ cylinder/2) myopiaandglaucoma,alsoshowninTable2.Glaucomawas measured in diopters (D). Myopia was defined when myopic almost threefold as frequent in eyes with moderate-to-high myopia spherical equivalent of the eye (SEq) was -1.0 D or greater, (4.4%) than in eyes without myopia (1.5%) (OR, 3.3; CI, 1.7–6.4) hyperopia if SEq was greater than ϩ1.0 D, and emmetropia when after adjusting for other known glaucoma risk factors. Again, it SEq was in the range from -0.99 to ϩ1.0 D. Low myopia was was consistently present across the age groups. defined in eyes with a myopic SEq of-1.0 D or greater to less than A weaker association was found between low myopia and OH, -3.0 D. Moderate-to-high myopia was defined in eyes with a asshowninTable3.TheOHwasmorefrequentineyeswithlow myopic SEq of-3.00 D or greater. For analyses involving persons, myopia (4.8%) than in eyes without myopia (2.9%) (OR, 1.7; CI, SEq was calculated as the mean value of the two eyes. 1.1–2.7) after adjusting for age and gender. This relationship Diabetes was diagnosed from history or an elevated fasting remained after simultaneously adjusting for other known glaucoma bloodglucoseof7.8mmol/lorgreater(140mg%).27Hyperten- risk factors as well (OR, 1.8; CI, 1.2–2.9). The magnitude of this

2011 Ophthalmology Volume 106, Number 10, October 1999

Table 1. Prevalence of Low and Moderate to Severe Myopia in Participant Eyes by Age Group and Sex, after Excluding Nonphakic Eyes

Women (3927 eyes) Men (2986 eyes) Either Sex (6913 eyes) Age Group (yrs) Myopic Eyes (%) Myopic Eyes (%) Myopic Eyes (%) Low Myopia (ՆϪ1.0 to ϽϪ3.0 D) Ͻ60 110 (10.4) 94 (11.4) 205 (10.9) 60–69 71 (5.4) 75 (6.8) 146 (6.0) 70–79 58 (5.8) 60 (8.6) 118 (7.0) 80ϩ 33 (10.1) 22 (10.1) 55 (10.0) All ages 272 (6.9) 252 (8.4) 524 (7.6) P for trend 0.04 0.21 0.02 Moderate to High Myopia (ՆϪ3.0 D) Ͻ60 81 (7.1) 47 (5.3) 128 (5.9) 60–69 75 (5.3) 31 (2.7) 106 (4.2) 70–79 39 (3.8) 38 (5.2) 77 (4.4) 80ϩ 19 (5.4) 12 (3.6) 31 (5.4) All ages 214 (5.4) 128 (4.3) 342 (4.9) P for trend 0.006 0.87 0.12 relationship, however, varied across the age groups examined, Table4showsthatthemeanIOPsofbothright(16.51mmHg) ranging from OR 0.5 to 3.9. No association was found between and left (16.40 mmHg) eyes with myopia were approximately a moderate-to-high myopia and OH. Similar prevalence rates for OH half-millimeter higher than emmetropic eyes (16.01 and 16.00 were found in eyes with moderate-to-high myopia (3.5%) to eyes mmHg, respectively), a small difference that was statistically sig- without myopia (2.9%) (OR, 0.9; CI, 0.4–2.0) after adjusting for niﬁcant by Student’s t test. The mean IOP for myopic eyes was other known glaucoma risk factors. 16.45 mmHg. Hyperopic eyes had a similar mean IOP (16.03 Associations between myopia and glaucoma or OH were also mmHg for the two eyes) as emmetropic eyes (16.00 mmHg). Eyes assessed in subjects in the multivariate model using mean refrac- with moderate-to-high myopia (Ն-3.00 D) also had a similar mean tive error of the two eyes and the presence of glaucomatous IOP of 16.47 mmHg to eyes with low myopia (Ն-1.00 D to Ͻ-3.0 damage or elevated IOP in either eye. For glaucoma, a similar D), which had a mean IOP of 16.44 mmHg. magnitude of association was found for subjects with any myopia The role of IOP in the association between glaucoma and myopia (SEq Ն -1.0 D) (OR, 2.4; CI, 1.5–4.0). Age-speciﬁc prevalence was also assessed by adjusting for IOP (continuously) in the multi- rates for glaucoma in subjects with and without myopia are com- variate model. The associations found with glaucoma were similar paredwithoverallglaucomaprevalencerates,asshowninFigure with OR of 2.1 (CI, 1.2–4.4) for low myopia and OR of 2.6 (CI, 1.Myopiawasassociatedwithasubstantiallyincreasedriskof 1.3–5.2) for moderate-to-high myopia. This suggests that the relation- glaucoma at all except the youngest age group. ship between glaucoma and myopia is independent of IOP.

Table 2. Associations between Glaucoma and Low Myopia (spherical equivalent ՆϪ1.0 to ϽϪ3.0 diopters) or Moderate to High Myopia (spherical equivalent ՆϪ3.00 diopters), Stratiﬁed by Age and Adjusted for Confounders

Age Glaucoma in Eyes Glaucoma in Eyes Multivariate- Group with Low Myopia without Myopia Sex-adjusted adjusted OR (yrs) [No. (%)] [No. (%)] OR (95% CI) (95%CI)* Ͻ60 1 (0.3) 4 (0.2) 1.7 (0.2–13.9) Insufﬁcient data 60–695(4.0)16(0.7)2.5(0.8–7.5)2.6(0.9–7.7) 70–79 7 (5.9) 41 (2.6) 2.1 (0.8–5.9) 2.1 (0.8–5.7) 80ϩ 9 (16.7) 28 (5.7) 2.4 (1.0–5.4) 2.6 (1.1–6.2) Allages22(4.2)89(1.5)2.1(1.2–3.8)†2.3(1.3–4.1)†

Glaucoma in Eyes with Glaucoma in Eyes Multivariate- Moderate to High without Myopia Sex-adjusted adjusted OR Myopia [No. (%)] [No. (%)] OR (95% CI) (95%CI)* Ͻ60 0 (0.0) 4 (0.2) Insufﬁcient data Insufﬁcient data 60–69 5 (4.7) 16 (0.7) 7.0 (2.0–24.8) 4.7 (1.0–22.6) 70–79 4 (5.3) 41 (2.6) 2.7 (1.0–7.4) 2.4 (0.9–6.8) 80ϩ 6 (19.4) 28 (5.7) 4.1 (1.4–12.1) 5.2 (1.7–15.8) Allages15(4.4)89(1.5)3.7(2.0–6.9)†3.3(1.7–6.4)†

OR ϭ odds ratio; CI ϭ conﬁdence interval. * Adjusted for sex, family history of glaucoma, diabetes, hypertension, typical migraine history, steroid use, and presence of pseudoexfoliation. † Adjusted also for age.

2012 Mitchell et al ⅐ Relationship between Glaucoma and Myopia

Table 3. Associations between Ocular Hypertension and Low Myopia (spherical equivalent ՆϪ1.0 to ϽϪ3.0 diopters) or Moderate to High Myopia (spherical equivalent ՆϪ3.00 diopters), Stratiﬁed by Age and Adjusted for Confounders

Age Ocular Hypertension in Ocular Hypertension in Multivariate- Group Eyes with Low Myopia Eyes without Myopia Sex-adjusted adjusted OR (yrs) [No. (%)] [No. (%)] OR (95% CI) (95% CI) Ͻ60 10 (4.9) 34 (2.0) 2.0 (0.7–4.2) 2.2 (1.0–4.5) 60–69 2 (1.4) 74 (3.2) 0.5 (0.1–2.1) 0.5 (0.2–1.2) 70–798(6.8)45(2.9)2.7(1.2–6.4)2.8(1.2–6.4) 80ϩ 5 (9.3) 21 (4.2) 2.0 (0.6–6.9) 3.9 (1.1–14.4) Allages25(4.8)174(2.9)1.7(1.1–2.7)†1.8(1.2–2.9)†

Ocular Hypertension in Ocular Hypertension in Multivariate- Eyes with Moderate to Eyes without Myopia Sex-adjusted adjusted OR High Myopia [No. (%)] [No. (%)] OR (95% CI) (95%CI)* Ͻ60 6 (4.7) 34 (2.0) 1.9 (1.3–2.7) 1.8 (0.5–6.3) 60–69 5 (4.8) 74 (3.2) 1.4 (0.5–3.9) 0.9 (0.3–2.7) 70–79 4 (5.3) 45 (2.9) Insufficient data Insufficient data 80ϩ 1 (1.3) 21 (4.2) Insufficient data Insufficient data Allages12(3.5)174(2.9)1.1(0.5–2.2)†0.9(0.4–2.0)†

* Adjusted for sex, family history of glaucoma, diabetes, hypertension, typical migraine history, steroid use, and presence of pseudoexfoliation. † Adjusted also for age.

Discussion damage.32ThelikelihoodthatsomepeoplewithOHhave an early stage of glaucoma is consistent with our finding of This population-based study of an Australian white com- a relatively weak association between OH and myopia, munity has found a strong relationship between open-angle which was only found with low myopia. This weak associ- glaucoma and myopia, after taking into account the effects ation probably is reflected by the slightly higher (half mil- of other known glaucoma risk factors. This association was limeter) IOPs of myopes. present for eyes with low myopia (OR, 2.3) and was stron- Although many clinic-based and case–control studies ger (OR, 3.3) for eyes with moderate-to-high myopia, sug- have suggested a relationship between glaucoma and myo- gesting a dose response. The consistency of these findings pia, no previous population-based studies have examined provides support for the hypothesis of a true relationship the association in detail while taking into account the effects between the two conditions. of other known glaucoma risk factors. Our findings are It may be deceptive to describe people with normal discs relevant, as the population-based design is likely to have and fields but elevated IOP by a different term (OH), as IOP minimized the possibility of selection bias affecting the elevation may have been present for too short a period and results. Confirmation that myopia is a frequent risk factor the visual field test may be too insensitive to detect early for glaucoma (one in four glaucoma cases in this age group) should help to identify this group, whose participants need earlier and regular ophthalmic screening and closer follow- up. Could our findings of a strong association between glaucoma and myopia be confounded by measurement error or misclassification of either the glaucoma or myopia status of persons in this older population? Such misclassification could bias the findings and lead to an overestimation of any relationship between the two conditions. The overall glaucoma prevalence reported from our study(3.0%)24ishigherthaninsomeotherrecentpopula- tion-basedprevalencestudies,includingBeaverDam32and Rotterdam.33However,thesetwostudiesexaminedeithera youngeragerange33orhadalowerproportionofsubjectsin theoldestagegroup.33Giventheexponentialincreasein glaucoma prevalence with age found in our study, such age differences could have a marked effect on the overall prevalence. Figure 1. Prevalence of glaucoma by age group in myopic and nonmyopic More likely sources of misclassification are the docu- subjects compared to prevalence in the overall Blue Mountains Eye Study mented changes in the appearance of the optic disc in population. myopia.34Thesecouldeitherresultintheoverclassification

2013 Ophthalmology Volume 106, Number 10, October 1999

Table 4. Mean Intraocular Pressure of Right and Left Eyes by Refractive Error Status, Using Spherical Equivalent Refraction (SEq) MeasuredinDiopters(D)*

Mean Intraocular Pressure Myopia SEq Emmetropia SEq ؊0.99 D Hyperopia SEq mmHg ؎ SE) >؊1.00 D P to ؉1.00 D P >؉1.00 D) Right eyes (n ϭ 3355) 16.51 (Ϯ0.13) 0 16.01 (Ϯ0.07) 0.22 16.14 (Ϯ0.07) Left eyes (n ϭ 3374) 16.40 (Ϯ0.13) 0 16.00 (Ϯ0.07) 0.46 15.93 (Ϯ0.07)

* Adjusted for age, sex, diabetes and systolic blood pressure, after excluding nonphakic eyes plus eyes with pseudoexfoliation and secondary or angle-closure glaucoma. of nonglaucomatous field defects resulting from tilted discs indicated that people with high myopia are more likely to be as glaucomatous or underclassification of glaucoma because steroidrespondersthanthoseinthegeneralcommunity.4 of the difficulties in grading the cup–disc ratio and neuro- Recently, the gene coding for a trabecular meshwork-in- retinal rim in some myopic eyes. Tilted discs may cause duced glucocorticoid response protein in the GLC1A locus visual field defects, but these are predominantly temporal on chromosome 1q21-q31 was identified and found in 3.9% andoftencrosstheverticalmeridian.35Patientswithmyo- of a glaucoma population compared to 0.3% of a general pia may also develop enlarged blind spots, superotemporal population.43Itispossiblethatthisandotherglaucoma defects,andirregulardefectsfrommyopicretinopathy.36 genes may be represented more frequently in persons with Great care was taken in the adjudication of field defects to myopia. includeonlythosedefectsthatweretypicallyglaucoma- In summary, these data from the Blue Mountains Eye tous.24Patientswithmyopiaalsotendtohavelargeroptic Study have confirmed the strong relationship between glau- discs34andcups.37Thestereoopticdiscphotographsfrom coma and myopia in an older white population sample. all myopic eyes classified as having glaucomatous field Subjects with myopia had a twofold to threefold increased defects were also carefully reviewed. Disc cupping, which risk of glaucoma compared with that of nonmyopic sub- matched the field defects, was confirmed in these glaucoma jects. This risk was independent of other glaucoma risk cases, although rim thinning was occasionally more difficult factors and IOP. Myopic eyes had slightly higher IOPs than to evaluate, because of the flatter nature of some myopic emmetropic or hyperopic eyes, and we found only a bor- discs. derline association between myopia and OH. Care was taken in the assessment of myopia to include only phakic eyes, as cataract surgery would affect measurement of the underlying refractive state. Although the myopia prevalence rates in this study are slightly lower than in References comparableUnitedStatesstudies,38,39thesedifferences may be due in part to small differences in myopia definition. 1. Knapp A. Glaucoma in myopic eyes. Arch Ophthalmol 1926; The 0.45-mmHg IOP difference found in our study be- 55:35–7. tween myopic and emmetropic eyes is slightly lower than in 2. Mo¨ller HU. Excessive myopia and glaucoma. Acta Ophthal- some previous reports, which have reported differences mol 1948;26:185–93. rangingfrom0.75tomorethan1.0mmHg.5,15,40Wewere 3. Fong DS, Epstein DL, Allingham RR. Glaucoma and myopia: also unable to confirm a trend with increasing levels of are they related? Int Ophthalmol Clin 1990;30:215–8. myopia,whichhasbeendemonstratedpreviously.17,40The 4. Podos SM, Becker B, Morton WR. High myopia and primary small difference found between myopic and nonmyopic open-angle glaucoma. Am J Ophthalmol 1966;62:1038–43. eyes,althoughstatisticallysignificant,isunlikelytobe 5. Perkins ES, Phelps CD. Open angle glaucoma, ocular hyper- important clinically. tension, low-tension glaucoma, and refraction. Arch Ophthal- mol 1982;100:1464–7. A number of mechanisms have been postulated to ex- 3 6. Phelps CD. Effect of myopia on prognosis in treated primary plainthelinkfoundbetweenglaucomaandmyopia. The open-angle glaucoma. Am J Ophthalmol 1982;93:622–8. optic nerve head in myopic eyes may be structurally more 7. Lotufo D, Ritch R, Szmyd L Jr, Burris JE. Juvenile glaucoma, susceptible to glaucomatous damage from elevated or nor- race, and refraction. JAMA 1989;261:249–52. malIOP5,7,9,34thaninnonmyopiceyes.Quigley31haspro- 8. Mastropasqua L, Lobefalo L, Mancini A, et al. Prevalence of posed that shearing forces exerted by scleral tension across myopia in open angle glaucoma. Eur J Ophthalmol 1992;2: the lamina cribrosa may be important in the pathogenesis of 33–5. glaucomatousdamage.CahaneandBartov41calculatedthat 9. Chihara E, Liu X, Dong J, et al. Severe myopia as a risk factor myopic eyes have higher scleral tension across the lamina for progressive visual field loss in primary open-angle glau- than in eyes with shorter axial length, even when IOP is the coma. Ophthalmologica 1997;211:66–71. 10. Daubs JG, Crick RP. Effect of refractive error on the risk of same, with the difference even more marked in eyes with ocular hypertension and open angle glaucoma. Trans Ophthal- thinner sclera. Similar connective tissue changes may also mol Soc U K 1981;101:121–6. 3,42 occuringlaucomaandmyopia. 11. Wilson MR, Hertzmark E, Walker AM, et al. A case-control Both glaucoma and myopia have a strong familial basis study of risk factors in open angle glaucoma. Arch Ophthal- and may share a common genetic link. An early report mol 1987;105:1066–71.

2014 Mitchell et al ⅐ Relationship between Glaucoma and Myopia

12. Charliat G, Jolly D, Blanchard F. Genetic risk factor in pri- 29. Cumming RG, Mitchell P, Leeder SR. Use of inhaled corti- mary open-angle glaucoma: a case–control study. Ophthalmic costeroids and the risk of cataracts. N Engl J Med 1997;337: Epidemiol 1994;1:131–8. 8–14. 13. Goldwyn R, Waltman SR, Becker B. Primary open-angle 30. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: glaucoma in adolescents and young adults. Arch Ophthalmol a generalized estimating equation approach. Biometrics 1988; 1970;84:579–82. 44:1049–60. 14. Tomlinson A, Phillips CI. Applanation tension and axial 31. Quigley HA. Reappraisal of the mechanisms of glaucomatous length of the eyeball. Br J Ophthalmol 1970;54:548–53. optic nerve damage. Eye 1987;1:318–22. 15. Abdalla MI, Hamdi M. Applanation ocular tension in myopia 32. Klein BE, Klein R, Sponsel WE, et al. Prevalence of glau- and emmetropia. Br J Ophthalmol 1970;54:122–5. coma. The Beaver Dam Eye Study. Ophthalmology 1992;99: 16. Seddon JM, Schwartz B, Flowerdew G. Case–control study of 1499–504. ocular hypertension. Arch Ophthalmol 1983;101:891–4. 33. Dielemans I, Vingerling JR, Wolfs RC, et al. The prevalence 17. David R, Zangwill LM, Tessler Z, Yassur Y. The correlation of primary open-angle glaucoma in a population-based study between intraocular pressure and refractive status. Arch Oph- thalmol 1985;103:1812–5. in The Netherlands. The Rotterdam Study. Ophthalmology 18. Linner E. The association of ocular hypertension with the 1994;101:1851–5. exfoliation syndrome, the pigmentary dispersion syndrome 34. Jonas JB, Gusek GC, Naumann GO. Optic disk morphometry and myopia. Surv Ophthalmol 1980;25:145–7. in high myopia. Graefes Arch Clin Exp Ophthalmol 1988;226: 19. Quinn GE, Berlin JA, Young TL, et al. Association of intraoc- 587–90. ular pressure and myopia in children. Ophthalmology 1995; 35. Brazitikos PD, Safran AB, Simona F, Zulauf M. Threshold 102:180–5. perimetry in tilted disc syndrome. Arch Ophthalmol 1990;108: 20. Ganley JP. Epidemiological aspects of ocular hypertension. 1698–700. Surv Ophthalmol 1980;25:130–5. 36. Greve EL, Furuno F. Myopia and glaucoma. Albrecht Von 21. Bonomi L, Mecca E, Massa F. Intraocular pressure in myopic Graefes Arch Klin Exp Ophthalmol 1980;213:33–41. anisometropia. Int Ophthalmol 1982;5:145–8. 37. Tomlinson A, Phillips CI. Ratio of optic cup to optic disc. In 22. Leighton DA, Tomlinson A. Ocular tension and axial length of relation to axial length of eyeball and refraction. Br J Oph- the eyeball in open-angle glaucoma and low tension glau- thalmol 1969;53:765–8. coma. Br J Ophthalmol 1973;57:499–502. 38. Wang Q, Klein BE, Klein R, Moss SE. Refractive status in the 23. Tielsch JM, Katz J, Quigley HA. Diabetes, intraocular pres- Beaver Dam Eye Study. Invest Ophthalmol Vis Sci 1994;35: sure, and primary open-angle glaucoma in the Baltimore Eye 4344–7. Survey. Ophthalmology 1995;102:480–53. 39. Sperduto RD, Seigel D, Roberts J, Rowland M. Prevalence of 24. Mitchell P, Smith W, Attebo K, Healey PR. Prevalence of myopia in the United States. Arch Ophthalmol 1983;101: open-angle glaucoma in Australia. The Blue Mountains Eye 405–7. Study. Ophthalmology 1996;103:1661–9. 40. Shiose Y, Kitazawa Y, Tsukahara S, et al. Epidemiology of 25. Leske MC, Connell AM, Wu SY. Risk factors for open-angle glaucoma in Japan: a nationwide glaucoma survey. Jpn J glaucoma. The Barbados Eye Study. Arch Ophthalmol 1995; Ophthalmol 1991;35:133–55. 113:918–24. 26. Attebo K, Mitchell P, Smith W. Visual acuity and the causes 41. Cahane M, Bartov E. Axial length and scleral thickness effect of visual loss in Australia. The Blue Mountains Eye Study. on susceptibility to glaucomatous damage: a theoretical model Ophthalmology 1996;103:357–64. implementing Laplace’s law. Ophthalmic Res 1992;24: 27. Mitchell P, Smith W, Chey T, Healey PR. Open-angle glau- 280–4. coma and diabetes: the Blue Mountains Eye Study, Australia. 42. Curtin BJ, Iwamoto T, Renaldo DP. Normal and staphyloma- Ophthalmology 1997;104:712–8. tous sclera of high myopia. An electron microscopic study. 28. Wang JJ, Mitchell P, Smith W. Is there an association between Arch Ophthalmol 1979;97:912–5. migraine headache and open-angle glaucoma? Findings from 43. Stone EM, Fingert JH, Alward WLM, et al. Identiﬁcation of a the Blue Mountains Eye Study. Ophthalmology 1997;104: gene that causes primary open angle glaucoma. Science 1997; 1714–9. 275:668–70.

2015