Risk factors for primary open-angle glaucoma in a Burmese population: the Meiktila Eye Study

Original Article

Risk factors for primary open-angle glaucoma in a Burmesepopulation: the Meiktila Eye StudyRobert J Casson FRANZCO,1 Aanchal Gupta MBBS,1 Henry S Newland FRANZCO,1 Steve McGovernFRANZCO,1 James Muecke FRANZCO,1 Dinesh Selva FRANZCO1 and Than Aung MD2

1South Australian Institute of Ophthalmology and Department of Ophthalmology and Visual Sciences, Adelaide University, Adelaide,South Australia, Australia; and 2Yangon Eye Hospital, Yangon, Myanmar

ABSTRACT

Purpose: To report the risk factors associated with primaryopen-angle glaucoma (POAG) in the Burmese population.

Methods: The Meiktila Eye study, a population-basedcross-sectional study, included inhabitants 40 years of ageand over from villages in the Meiktila District. Of 2481eligible participants identified, 2076 participated in the studyand sufficient examination data to diagnose glaucoma in atleast one eye was obtained in 1997 participants. Theophthalmic examination included slit-lamp examination,tonometry, gonioscopy and dilated stereoscopic fundusexamination. Definitions adhered to the InternationalSociety for Geographic and Epidemiological Ophthalmolo-gy’s recommendations. Univariate and multivariate analysesof potential risk factors were performed.

Results: The overall prevalence of POAG was 2.0% (95% CI0.9–3.1). In the univariate analysis, increasing age (P = 0.024),spherical equivalent (P = 0.01), axial length (P = 0.023) andintraocular pressure (IOP; P < 0.001) were significantly asso-ciated with POAG. And in the multivariate analysis, myopia<0.5 D (P = 0.049), increasing age and IOP (P < 0.001) weresignificant risk factors for POAG.

Conclusion: POAG in this Burmese population was associ-ated with increasing age, axial myopia and IOP.

Key words: glaucoma, open-angle, prevalence, risk factor,rural population.

INTRODUCTION

Glaucoma is the second leading cause of blindness in theworld, with open-angle glaucoma (OAG) the most prevalentform.1 Quigley and Broman estimate that worldwide nearly44.7 million people will have primary open-angle glaucoma(POAG) by the year 2010, and 4.5 million will be blinded byit.1 The prevalence of POAG varies among different racialgroups. Rates reported in white adult populations in theBeaver Dam and Blue Mountains Eye studies were 2.1% and3.0%, respectively;2,3 in black populations, the prevalence is2–3 times higher;4–6 the prevalence in Indian, Japanese andSouth-east Asian populations appears similar to that in whitepopulations.7–11

In terms of risk factors, intraocular pressure (IOP)12,13

age,3,4,9,14,15 positive family history,16–20 myopia4,8,21–23 and lowperfusion pressure24 have been convincingly associated witha higher incidence of POAG. A number of other risk factorsfor POAG have been inconsistently reported, includingdiabetes mellitus, hypertension, smoking, migraine, thyroiddisease and alcohol consumption.8,16–18,22,25–31 To date, limiteddata from Myanmar have been available on the prevalence ofglaucoma. This study aims to report the risk factors associ-ated with POAG in the Burmese population.

METHODS

Sampling procedure

The Meiktila Eye Study (MES) was a population-based,cross-sectional ophthalmic survey of the inhabitants of ruralvillages in central Myanmar. The principle aims of thisproject were to estimate the prevalence and causes of visual

� Correspondence: Associate Professor Robert Casson, South Australian Institute of Ophthalmology, Department of Ophthalmology and Visual Sciences,

Adelaide University, South Australia 5000, Australia. E-mail: [email protected]

Received 26 May 2007; accepted 21 August 2007.

Clinical and Experimental Ophthalmology 2007; 35: 739–744doi: 10.1111/j.1442-9071.2007.01619.x

© 2007 The AuthorsJournal compilation © 2007 Royal Australian and New Zealand College of Ophthalmologists

mailto:[email protected]

impairment and the prevalence and risk factors of oculardisorders, including glaucoma, among persons 40 years ofage or older in this region.

The study was conducted within the Mandalay Division,an area encompassing 34253 km2 divided into seven second-order administrative districts of approximately equal size.The township of Meiktila (population approximately251 000) located at 20°53′N, 95°53′E lies centrally in theMeiktila District, and is the only urban region in this entireDistrict. The District is arbitrarily divided by the Ministry ofHealth (MOH) into six zones served by a centrally locatedeye hospital in Meiktila.

Participants were randomly selected using a stratified,cluster sampling process. A sampling frame consisting ofthe list of all villages in the Meiktila District with theirpopulations was obtained from the MOH. Villages werearbitrarily stratified as large (population >825) or small(population �825), with small villages in each of the sixzones within the Meiktila District constituting six separatestrata. For logistical reasons, sampling was restricted to vil-lages within 3 h drive from Meiktila (an area encompassingapproximately 80% of the District). All persons 40 yearsand over within each selected village were eligible forinclusion. The sample size was based on the desired preci-sion of the estimate of blindness (the principal aim of theMES); the assessment of glaucoma prevalence was a sec-ondary objective. Health-care workers from Meiktila town-ship enumerated the selected villages (and advertised andpromoted the survey) prior to commencement of thesurvey. Six small villages (one from each zone) and fourlarge villages were enumerated, providing a total samplepopulation of 2481 people.

Data collection

Data collection occurred at the end of the rainy season inNovember 2005. A single survey team conducted the entirestudy. Each team member was assigned specific tasks and waswell trained in the appropriate area. Specific observationswere performed by 1–2 members, limiting or eliminatinginter-observer variability. All equipment and personnel weretransported to each village, and the data collection occurredon site. A medical and ophthalmic history was obtained fromeach patient in their own language by qualified health-careworkers. Each participant then received a comprehensivevision and eye examination.

Intraocular pressure was measured with a Goldmannapplanation tonometer (Haag-Streit, Köniz, Switzerland)and anterior segment examination was performed at the slitlamp (Haag-Streit). The presence of previous iris ischaemiaor pseudoexfoliation was recorded. Gonioscopy was per-formed by two experienced ophthalmologists (RJC and SM)using a Sussman goniolens. Static gonioscopy was performedin dim illumination with minimal pressure on the corneausing a short slit beam; each quadrant was graded using theScheie classification. If >90° of posterior trabecular mesh-work (posterior TM) was visible the pupil was dilated with

tropicamide 1% and phenylephrine 2.5%. Eyes with �90°of posterior TM visible were deemed ‘occludable’, as perInternational Society for Geographic and EpidemiologicalOphthalmology (ISGEO) guidelines, and were dilated withtropicamide 0.5% only and kept under observation for 4 h(the maximum practicable period). If not possible, they werenot dilated. If either eye had evidence of previous acuteangle-closure glaucoma (see later definition) then neither eyewas dilated. Optic disc and retinal examination was per-formed by two experienced ophthalmologists (HSN and JM)using a 78 D lens and reference to standard disc images. Thevertical cup/disc ratio (CDR) and the presence of focalnotching were recorded.

As established during the pre-survey training period, theagreement between the two ophthalmologists for gradingthe angle was good (kappa = 0.78, as was determining theCDR kappa = 0.72).

Eyes with visual acuity (VA) > 6/60 and which fulfilledCategory 1 optic disc criteria (Table 1) underwent fullthreshold perimetry (C-20 strategy) using Frequency Dou-bling Technology (Zeiss Humphrey Systems, Dublin, CA,USA). Tests were considered reliable if there were less than20% fixation errors and less than 33% false-positive andfalse-negative errors. All individuals were naïve to perim-etry and received instruction in their own language fol-lowed by a practice in the demonstration mode. If theinitial test was unreliable, individuals were given a secondattempt. If patients did not perform reliable visual fieldsafter a second attempt they were excluded. More than 1missed point on the pattern deviation was consideredabnormal.

Definitions

In the MES two definitions of blindness were used. Presentingblindness was defined as unaided VA (or with spectacles ifworn) <6/120 in the better eye; and WHO blindness wasdefined as corrected VA <6/120 in the better eye. Fielddefects were not taken into consideration.32

A three-tired system of evidence, as suggested by theISGEO was used to categorize glaucoma (Table 1).

Prevalence was calculated on an individual rather than aper-eye basis. If at least one eye was diagnosable using theabove criteria, the subject was included in the prevalenceanalysis. In those subjects with only one diagnosable eye, ifglaucoma was not present in this eye, they were assumed notto have glaucoma in the un-diagnosable eye.

Primary open-angle glaucoma was diagnosed if the crite-ria for categories 1–3 were met and >90° of posterior TM wasvisible on static gonioscopy and no secondary cause forglaucoma was present.

Acute angle-closure glaucoma was diagnosed if the crite-ria for categories 1–3 were met, �90° of posterior TM wasvisible with static gonioscopy, no secondary cause for glau-coma was present and there was evidence of previous irisischaemia (defined as the presence of iris whorling or stromalatrophy).32

740 Casson et al.


Statistics

Prevalence rates were calculated as ratio estimates usingappropriate weights for each of the sampled villages. Boot-strapping was used to overcome the problem of varianceestimation in clusters where only the one primary samplingunit (village) was selected. All prevalence estimates wereperformed using SAS Version 9.1 (SAS Institute Inc., Cary,NC, USA).

Univariate analyses were performed using single variablebinary logistic regression (equivalent to t-tests for continuousor chi-squared tests for categorical data). A multivariateregression model was constructed. To reduce the likelihoodof ‘over-fitting’ the model, the ratio of the number of subjectswith the condition of interest (the dependent variable) to thenumber of predictors was 10:1 and the predictors werechosen before data exploration, and in order were: IOP,myopia, age and axial length. To preserve information integ-rity, continuous variables were not categorized for themultivariate analysis and forced entry method was used.Residuals were examined for influential cases using Cook’sdistance and tests, including average variance inflation factorfor the presence of multicollinearity were performed. Statis-tical analysis was performed using a commercially availablestatistical software package (SPSS for Windows, version 11.5,SPSS Inc., Chicago, IL, USA). Odds ratios (ORs) with 95%confidence intervals (CIs) were calculated for each predictor.A Hosmer and Lemeshow goodness-of-fit and a NagelkerkeR2 were calculated. In subjects positive for the dependentvariable in both eyes, the right eye was used for analysis, andin unilateral cases, all variables were drawn from the affectedside. A P-value of <0.05 was considered significant.

RESULTS

A total of 2481 subjects were eligible and 2076 were exam-ined (participation rate 83.7%). The mean age � standarddeviation was 56.2 � 12 years. The mean IOP in the righteye was 14.8 mmHg (95% CI 14.65–14.95). Sufficientexamination data to diagnose glaucoma (as defined above) inat least one eye was obtained in 1997 participants of whom40 had POAG. The overall prevalence of POAG was 2.0%(95% CI 0.9–3.1). Table 2 shows the univariate analysis forOAG. Increasing age (OR 2.70, 95% CI 1.15–6.37,P = 0.024 for the 70+ years age group), myopia (OR 2.82,95% CI 1.28–6.25, P = 0.01), IOP (OR 1.06, 95% CI 1.03–1.09, P < 0.001) and axial length (OR 1.36, 95% CI 1.01–1.77, P = 0.023) were significantly associated with POAG.Other variables such as smoking, illiteracy, gender andaverage radius of corneal curvature did not show a statisti-cally significant association with POAG.

In the multivariate analysis (Table 3), myopia (OR 2.74,95% CI 1.0–7.48, P = 0.049), increasing age and IOP (OR1.14, 95% CI 1.07–1.21, P < 0.001) were significant riskfactors for OAG. The association between axial length andage with POAG, while present, failed to reach statisticalsignificance.

DISCUSSION

The MES provides the first population-based data about theprevalence of POAG in Myanmar. The overall prevalence ofPOAG was 2.0%. The age-adjusted prevalence of glaucomain the MES has been reported elsewhere.33 This is compa-rable to the prevalence of POAG in other reported South-east Asian and white populations, where the prevalenceestimates vary from 0.5% to 2.5%2,3,7–11,14,15,23; but lower thanrates reported in black populations.4–6

In the present study, subjects aged 70 years or more were2.7 times more likely to have POAG than those in their fifthdecade, a finding consistent with other reports regardless ofthe ethnicity of the participants.3,4,7,9,10,15,23 Reports of anassociation between gender and POAG are inconsistent;some studies find a higher incidence in men,4,7,15 others inwomen3 and some found no difference in gender preva-lence.2,8,23 We found no significant gender difference in thisBurmese population.

Myopia is consistently reported as a risk factor forPOAG.4,8,21–23 In the present study, the OR in the multivariateanalysis was 2.74 (95% CI, 1.0–7.48) for myopes with aspherical equivalent of <-0.5 D. The OR was higher than thatfound in the Barbados Eye study (OR 1.5, 95% CI 1.1–2.0),4

but similar to that found in the Tajimi study (OR 1.85, 95% CI1.03–3.31) for low myopes,8 the Aravind Comprehensive EyeSurvey (OR 2.9, 95% CI 1.3–6.9) for mild myopia,7 and theBlue Mountains Eye Study (OR 2.3, 95% CI 1.3–4.1).22

Although myopia has been identified as a risk factor forPOAG (independent of age and IOP in diverse populations),the pathophysiological mechanism-linking myopia andPOAG remains unclear. A commonly held view is that the

Table 1. Diagnostic criteria for glaucoma35

Category 1 diagnosis (structural and functional evidence): Eyes witha CDR > 97.5th percentile for the normal (non-glaucomatous)population (CDR � €0.7 was used based on data from previ-ous studies in the region) or a CDR � €0.6 in the presence ofasymmetry �0.3 or a neuroretinal rim width reduced to <0.1CDR (between 11 and 1 o’clock or 5–7 o’clock) plus a definitevisual field defect consistent with glaucoma. Eyes with evi-dence of previous AACG which were no perception of lightwere also classified as Category 1 even if the optic disc was notvisualized (‘end stage’ AACG).

Category 2 diagnosis (advanced structural damage with unprovedfield loss): If the subject could not satisfactorily complete visualfield testing but had a CDR > 99.5th percentile for thenormal (non-glaucomatous) population (CDR � €0.8 was usedbased on data from the normal population in this study) glau-coma was diagnosed solely on the structural evidence.

Category 3 diagnosis (Optic disc not seen. Field test impossible) Ifit was not possible to examine the optic disc, glaucoma wasdiagnosed if: (A) The visual acuity was <6/120 and the IOP>99.5th percentile or (B) The visual acuity was <6/120 and theeye showed evidence of glaucoma filtering surgery.

AACG, acute angle-closure glaucoma; CDR, vertical cup/discratio; IOP, intraocular pressure.

Risk factors for primary open-angle glaucoma 741


larger eye and/or thinner scleral coat cause the optic nerveaxons at the lamina cribrosa to be particularly susceptibleto IOP.34 Hence, it is axial myopia which is assumed to bethe culprit; however, population-based data supportingthis hypothesis is scarce. In the current study, the univariateand multivariate analyses taken together suggest that axialmyopia is associated with POAG in this population.

Intraocular pressure was the most significant risk factor forPOAG in the present study, with a similar OR (1.14) to thatfound in other Asian populations: 1.13 (95% CI 1.04–1.21)in the Tajimi study8 and 1.12 (95% CI 1.08–1.16) in the ruralSouth Indian population of Tamil Nadu.23

The current study had some limitations. Of those eyesthat were not adequately examined for glaucoma, most haddense cataracts. It is possible that some of these eyes also hadglaucoma so that the overall estimate for the prevalence ofPOAG (2.0%) may, in fact, be conservative. The 97.5thpercentile for the CDR in the normal population wasapproximately 6.5; however, based on recent data from thisregion a CDR of �0.7 was chosen as the cut-off for fieldtesting; hence, it is possible that early glaucoma with con-centric cupping was missed.

Although the participation rate was relatively high (83%),we have no robust data about the visual status and ocularhealth of the non-participants. Anecdotally (according to thevillage chiefs), the principal reason for non-participation wasoccupation-related; hence, it is unlikely that any of the non-participants were glaucoma blind, suggesting that the preva-lence of glaucoma in this group would be lower than in theparticipants. Although accurate data about the gender distri-bution in the Meiktila District was not available, it is likelythat women were over-represented in this study (59%), acommon occurrence among similar studies, possibly reflect-ing occupation-related availability.

The diagnosis of glaucoma relied on an accurate clinicalassessment of the optic disc by experienced ophthalmolo-gists who had access to all the clinical information. Theagreement between this method of assessment comparedwith disc assessment of photographs by masked observers isunclear.

The Meiktila District was chosen for logistical reasons, notrandomly, and may not be representative of neighbouringregions within the Mandalay Division of central Myanmar;however, we have no reason to believe that this is the case.

Table 2. Univariate risk factors for primary open-angle glaucoma

Odds ratio 95% confidenceinterval

P-value

Lower Upper

Age 1.03 1.0 1.06 0.013Age group (years)

40–49 1.0 1.0 1.050–59 0.84 0.31 2.25 0.7260–69 2.17 0.93 5.13 0.07770+ 2.70 1.15 6.37 0.024

GenderFemale 1.0 1.0 1.0Male 1.52 0.35 1.21 0.18

Illiteracy 1.2 0.42 1.59 0.57Smoking 1.27 0.64 2.50 0.49Spherical equivalent 1.12 1.03 1.22 0.011Myopia† 2.82 1.28 6.25 0.01Axial length 1.36 1.01 1.77 0.023Average radius corneal curvature 0.94 0.96 1.11 0.46Intraocular pressure 1.06 1.03 1.09 <0.001

†Myopia defined as spherical equivalent <-0.5 D.

Table 3. Multivariate risk factors for primary open-angle glaucoma

Odds ratio 95% confidenceinterval

P-value

Lower Upper

Intraocular pressure 1.14 1.07 1.21 <0.001Myopia† 2.74 1.0 7.48 0.049Age 1.03 1.0 1.6 0.044Axial length 1.90 0.80 4.57 0.15

Hosmer and Lemeshow goodness-of-fit (P = 0.76); Nagelkerke R2 (0.10) average variance inflation factor 1.04. †Myopia defined as sphericalequivalent <-0.5 D.

742 Casson et al.


In conclusion, the prevalence of POAG in the population40 years of age and over in the Meiktila District of rural,central Myanmar is 2.0%. It is associated with increasing age,myopia and IOP. These results are comparable to otherpopulation-based studies in other Asian population and somewhite populations.2,3,7–11,14

ACKNOWLEDGEMENTS

The survey was funded by a grant from Pfizer Australia, butthe design of the survey, its execution, analysis, interpreta-tion and publication were carried out independently by theauthors.

The MES was approved by the MOH in Myanmar andhad ethical approval from the Royal Adelaide HospitalEthics Committee. Consent for participation was obtainedfrom the head of each village prior to commencement of thesurvey; and written, informed consent, in the participantsown language, was obtained from all willing participants.The study was conducted in accordance with the Declara-tion of Helsinki.

REFERENCES

1. Quigley HA, Broman AT. The number of people with glau-coma worldwide in 2010 and 2020. Br J Ophthalmol 2006; 90:262–7.

2. Klein BE, Klein R, Sponsel WE et al. Prevalence glaucoma. TheBeaver Dam Eye Study. Opthalmology 1992; 99: 1499–504.

3. Mitchell P, Smith W, Attebo K, Healey PR. Prevalence ofopen-angle glaucoma in Australia. The Blue Mountains EyeStudy. Opthalmology 1996; 103: 1661–9.

4. Leske MC, Connell AM, Schachat AP, Hyman L. The BarbadosEye Study. Prevalence of open angle glaucoma. Arch Ophthalmol1994; 112: 821–9.

5. Mason RP, Kosoko O, Wilson MR et al. National survey of theprevalence and risk factors of glaucoma in St. Lucia, WestIndies. Part I. Prevalence findings. Ophthalmology 1989; 96:1363–8.

6. Tielsh JM, Sommer A, Katz J, Royall RM, Quigley HA, Javitt J.Racial variations in the prevalence of primary open-angle glau-coma. The Baltimore Eye Survey. JAMA 1991; 266: 369–74.

7. Ramakrishnan R, Nirmalan PK, Krishnadas R et al. Glaucoma ina rural population of southern India: the Aravind comprehen-sive eye survey. Opthalmology 2003; 110: 1484–90.

8. Suzuki Y, Iwase A, Araie M et al. Risk factors for open-angleglaucoma in a Japanese population: the Tajimi Study. Ophthal-mology 2006; 113: 1613–17.

9. Bourne RR, Sukudom P, Foster PJ et al. Prevalence of glaucomain Thailand: a population based survey in Rom Klao District,Bangkok. Br J Ophthalmol 2003; 87: 1069–74.

10. Foster PJ, Oen FT, Machin D et al. The prevalence of glaucomain Chinese residents of Singapore: a cross-sectional populationsurvey of the Tanjong Pagar district. Arch Ophthalmol 2000; 118:1105–11.

11. Rahman MM, Rahman N, Foster PJ et al. The prevalence ofglaucoma in Bangladesh: a population based survey in Dhakadivision. Br J Ophthalmol 2004; 88: 1493–7.

12. The Advanced Glaucoma Intervention Study (AGIS). 7. Therelationship between control of intraocular pressure and visualfield deterioration. The AGIS Investigators. Am J Ophthalmol2000; 130: 429–40.

13. Leske MC, Heijl A, Hussein M, Bengtsson B, Hyman L,Komaroff E. Factors for glaucoma progression and the effect oftreatment: the early manifest glaucoma trial. Arch Ophthalmol2003; 121: 48–56.

14. Bonomi L, Marchini G, Marraffa M et al. Prevalence of glau-coma and intraocular pressure distribution in a definedpopulation. The Egna-Neumarkt Study. Ophthalmology 1998;105: 209–15.

15. Dielemans I, Vingerling JR, Wolfs RC, Hofman A, GrobbeeDE, de Jong PT. The prevalence of primary open-angle glau-coma in a population-based study in the Netherlands. TheRotterdam Study. Ophthalmology 1994; 101: 1851–5.

16. Leske MC, Connell AMD, Wu SY, Hyman LG, Schachat AP.Risk factors for open-angle glaucoma. The Barbados Eye Study.Arch Ophthalmol 1995; 113: 918–24.

17. Nemesure B, Leske MC, He Q, Mendell N. Analyses ofreported family history of glaucoma: a preliminaryinvestigation. The Barbados Eye Study Group. Ophthalmic Epi-demiol 1996; 3: 135–41.

18. Tielsch JM, Katz J, Sommer A, Quigley HA, Javitt JC. Familyhistory and risk of primary open angle glaucoma. TheBaltimore Eye Survey. Arch Ophthalmol 1994; 112: 69–73.

19. Wilson MR, Hertzmark E, Walker AM, Childs-Shaw K, EpsteinDL. A case–control study of risk factors in open angleglaucoma. Arch Ophthalmol 1987; 105: 1066–71.

20. Wolfs RC, Klaver CCW, Ramrattan RS, van Duijn CM,Hofman A, de Jong PT et al. Genetic risk of primary open-angleglaucoma. Population-based familial aggregation study. ArchOphthalmol 1998; 116: 1640–5.

21. Mayama C, Suzuki Y, Araie M et al. Myopia advanced-stageopen-angle glaucoma. Ophthalmology 2002; 109: 2072–7.

22. Mitchell P, Hourihan F, Sandbach J, Wang JJ. The relationshipbetween glaucoma and myopia: the Blue Mountains Eye Study.Ophthalmology 1999; 106: 2010–15.

23. Vijaya L, George R, Paul PG et al. Prevalence of open-angleglaucoma on a rural south Indian population. Invest OphthalmolVis Sci 2005; 46: 4461–7.

24. Tielsch JM, Katz J, Sommer A, Quigley HA, Javitt JC. Hyper-tension, perfusion pressure, and primary open-angle glaucoma.A population-based assessment. Arch Ophthalmol 1995; 113:216–21.

25. Daubs JG, Crick RP. Effect of refractive error on the risk ofocular hypertension and open angle glaucoma. Trans OphthalmolSoc UK 1981; 101: 121–6.

26. Dielemans I, Vingerling JR, Algra D, Hofman A, Grobbee DE,de Jong PT. Primary open-angle glaucoma, intraocular pressure,and systemic blood pressure in the general elderly population.The Rotterdam Study. Ophthalmology 1995; 102: 54–60.

27. Katz J, Sommer A. Risk factors for primary open angleglaucoma. Am J Prev Med 1988; 4: 110–14.

28. Klein BEK, Klein R, Jensen SC. Open-angle glaucoma andolder-onset diabetes. The Beaver Dam Eye Study. Ophthalmol-ogy 1994; 101: 1173–7.

29. Klein BEK, Klein R, Ritter LL. Relationship of drinking alcoholand smoking to prevalence of open-angle glaucoma. The BeaverDam Eye Study. Ophthalmology 1993; 100: 1609–13.

Risk factors for primary open-angle glaucoma 743


30. Tielsch JM, Katz J, Sommer A, Quigley HA, Javitt JC. Hyper-tension, perfusion pressure, and primary open-angle glaucoma.A population-based assessment. Arch Ophthalmol 1995; 113:216–21.

31. Tielsch JM, Katz J, Quigley HA, Javitt JC, Sommer A. Diabe-tes, intraocular pressure, and primary opn-angle glaucoma inthe Baltimore Eye Survey. Ophthalmology 1995; 102: 48–53.

32. Casson RJ, Newland H, Muecke J et al. Prevalence and Causesof Visual Impairment in Rural Myanmar: The Meiktila EyeStudy. Ophthalmology 2007; April 18 [Epub ahead of print].

33. Casson RJ, Newland H, Muecke J et al. Prevalence of glaucomain rural Myanmar: the Meiktila Eye Study. Br J Ophthalmol 2007;91: 710–14.

34. Jonas JB, Berenshtein E, Holbach L. Lamina cribrosa thicknessand spatial relationships between intraocular space and cere-brospinal fluid space in highly myopic eyes. Invest Ophthalmol VisSci 2004; 45: 2660–5.

35. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The defini-tion and classification of glaucoma in prevalence surveys. Br JOphthalmol 2002; 86: 238–42.

744 Casson et al.


Documents

Risk factors for primary open-angle glaucoma in a Burmese population: the Meiktila Eye Study