You are here

Article Text

Article menu

Download PDFPDF

Global issues
Refractive error, axial length and anterior chamber depth of the eye in British adults: the EPIC-Norfolk Eye Study
  1. P J Foster1,2,
  2. D C Broadway3,
  3. S Hayat4,
  4. R Luben4,
  5. N Dalzell4,
  6. S Bingham5,
  7. N J Wareham4,
  8. K-T Khaw4
  1. 1Division of Genetics & Epidemiology, UCL Institute of Ophthalmology, University College London, UK
  2. 2National Biomedical Research Centre for Ophthalmology, Moorfields Eye Hospital, London, UK
  3. 3Department of Ophthalmology, Norfolk and Norwich University Hospital, Norwich, UK
  4. 4Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge School of Clinical Medicine, Cambridge, UK
  5. 5MRC Dunn Human Nutrition Unit, Cambridge University, Cambridge, UK
  1. Correspondence to Paul J Foster, Division of Genetics and Epidemiology, UCL Institute of Ophthalmology, 11-43 Bath Street, London EC1V 9EL, UK; p.foster{at}ucl.ac.uk

Abstract

Purpose To describe the distribution, and demographic and socioeconomic correlates of refractive error and related ocular biometry in an older British population.

Methods Refractive error was measured using an auto-refractor without cycloplegia. Pseudophakic individuals and those who had undergone refractive surgery were excluded from analysis. Axial length and anterior chamber depth were measured using partial coherence laser interferometry. Occupation category and highest educational achievement were recorded.

Results Biometric data were available for 2519 people (1090 men, 1429 women; 93.2% of all participants) aged 48 to 88 years. Refractive data were available for both eyes in 2210 bilaterally phakic participants. Among phakic individuals, axial length of the eye was strongly inversely correlated with refractive error in both men and women (p<0.001). Axial length of the eye was strongly, independently related to height, weight and social class, but most strongly related to educational achievement. In contrast, anterior chamber depth varied with age and sex, but not with socioeconomic status. There was a significant inverse association between anterior chamber depth and refraction in women (p<0.001) but not in men (p=0.495).

Conclusion Refractive error in this predominantly white older UK population was associated with axial biometry and sociodemographic characteristics. Educational status was the strongest determinant of axial length.

  • Anterior chamber
  • optics and refraction
  • epidemiology
  • anatomy

https://doi.org/10.1136/bjo.2009.163899

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Extremes of ocular dimensions are related to several pathological ophthalmic conditions, notably refractive error and glaucoma, which cause a large burden of ocular morbidity. Refractive error is the commonest eye disorder worldwide. Myopia prevalence tends to be lowest among people of European ancestry, with prevalence ranging from 12% to 26.5% among adults aged 40 years and older. Another 6–12% have a hypermetropic refraction.1–4 The myopia prevalence figures equate to an estimated 80 million people in Western Europe and the USA with a myopic refractive error (less than −1 D).5 The highest rates of myopia are found in industrialised regions of East Asia. Among Singaporean adults aged ≥40 years, the overall prevalence of myopia, hyperopia, astigmatism and anisometropia has been reported to be 39%, 28%, 38% and 16%, respectively, although this study used more liberal criteria for determining refractive error than the European review cited above (less than −0.5 D and more than +0.5 D for myopia and hypermetropic error, respectively).6 In Taipei, Taiwan, older people (≥65 years of age) have been found to have lower rates of myopia (less than −1.0 D, 14%), but more hypermetropia (>+1.0 D, 44%), astigmatism (cylinder >1.0 D, 45%) and anisometropia (difference >1.0 D, 22%) than Singaporeans.7 However, nationwide survey data have shown that younger Taiwanese people have extraordinarily high rates of myopia that, for those aged between 16 and18 years, have increased from 74% in 1983 to 84% in 2000.8

    Glaucoma is the leading cause of irreversible blindness worldwide.9 By 2010, an estimated 60.5 million people will have glaucomatous optic neuropathy, and 8.4 million will be blind in both eyes.10 Although open-angle glaucoma affects three times more people than angle-closure glaucoma (ACG) (predicted to affect 44.7 vs 15.7 million by 2010), there is an approximate 50:50 split in terms of the number of people blind in both eyes from the open-angle and angle-closure forms of glaucoma.10 Myopic refractive error has long been recognised as a risk factor for open-angle glaucoma.11 12 The association between open-angle glaucoma and refractive error appears to be driven by variation in ocular axial length (AL) (PJF. ARVO abstract, 2008). Conversely, a shallow anterior chamber and shorter AL of the eye are risk factors in the development of ACG.13 14

    The distribution of ocular biometric characteristics associated with refractive error and glaucoma is now reasonably well documented in Asian populations.15–19 However, published data are scarce for European-derived populations. In this report we examined the prevalence of refractive errors, as well as the distribution, associations and determinants of two important ocular biometric risk factors (AL and anterior chamber depth (ACD)).

    Methods

    The European Prospective Investigation of Cancer (EPIC) was conceived as a pan-European study of the genetic and environmental determinants of cancer20 and later expanded to study other health end points in older age. The EPIC-Norfolk cohort comprised 25 000 men and women aged 40–79 recruited between 1993 and 1997. Extensive lifestyle and biological data have been recorded in this longitudinal study of the determinants of health and disease in older people.21 A third health examination was initiated in 2006 that aimed to assess objectively physical, cognitive and ocular characteristics of participants now aged 48–89 years. The third health check was reviewed and approved by the East Norfolk and Waveney NHS Research Governance Committee (2005EC07L). Sex and age at time of examination were recorded. Social class was classified according to the Registrar General's occupation-based system, identifying five main categories with class I representing professionals, class II managerial and technical occupations, class III subdivided into non-manual and manual skilled workers, class IV partially skilled workers and class V unskilled manual workers. Social classes I, II and III non-manual were classified as non-manual workers, whereas social classes III-manual, IV and V were classed as manual workers. Educational attainment was based on highest qualification attained and was categorised into four groups: degree or equivalent, A level or equivalent, O level or equivalent, and less than O level or no qualifications. Based on previous analyses of the association between health and educational attainment, it was expected that the crucial cut-point would occur at the O level or better versus less than O level or no qualifications. Participant height and weight were measured when wearing light clothing and no shoes by trained nurses according to standard protocols. Height was measured to 0.1 cm using a stadiometer, whereas weight was measured to the nearest 0.1 kg using digital weighing scales (Tanita UK Ltd., Middlesex, UK).

    Refractive error was measured once in each eye using an auto-refractor (Humphrey model 500, Humphrey Instruments, San Leandro, California, USA). Pharmacological cycloplegia was not performed in view of the older age of participants. Pseudophakic participants and those who had undergone refractive surgery were excluded from analysis of refractive error. Both AL and ACD were measured using a Zeiss IOLMaster Optical Biometer (IOLMaster, Carl Zeiss Meditech Ltd, Welwyn Garden City, UK). Five measurements per eye were made of AL using the technique of partial coherence interferometry, and these were used to produce a mean value for each eye. Five measurements of ACD (including central corneal thickness) were made for each phakic eye using an optical technique. The ACD data from aphakic and pseudophakic eyes were excluded. Where data for both eyes were available, mean values for the two eyes were used as the measure for the individual. Where a measurement was only available for one eye, this was accepted as the measure for the individual. All ocular examinations were carried out according to Moorfields Eye Hospital standard operating procedures adapted for EPIC-Norfolk. Intensive periods of staff training and validation were conducted before the project commenced. Refresher training was provided annually during the data collection period. Data were reviewed weekly by a consultant ophthalmologist. Direct personal feedback was given to staff responsible for data collection.

    Statistical software (SPSS 11.5, Chicago, Illinois, USA) was used to calculate mean values and standard deviations of variables of interest. Differences between means were examined using the t test or analysis of variance methods. Differences in proportions were examined using the χ2 test. In addition, linear regression models were used for multivariate analyses examining the relationship between biometric covariates and age, sex, education, occupational class, height and weight.

    Results

    Biometric data were available for 2519 people (1090 men, 1429 women) aged 48 to 88 years. Both AL and ACD showed a normal distribution and were highly correlated with each other (r=0.43 in men, r=0.48 in women, p<0.0001). Among phakic individuals, AL was strongly inversely correlated with refractive error in both men and women (p<0.001 for both). There was a similar significant inverse association between ACD and refraction in women (p<0.001) but not in men (p=0.495). Table 1 provides mean age- and sex-specific refraction, AL and ACD. Mean AL (23.80 vs 23.29 mm, p<0.001) and ACD (3.15 vs 3.08 mm, p<0.001) were significantly greater in men compared to women. Mean AL was significantly longer in younger people than the more older of both sexes (men p<0.01, women p<0.001). There was also an inverse relationship between ACD and age, although the trend was not consistent in the oldest age group (aged ≥80 years) for whom ACD seemed to be deeper. No difference in mean refractive error was identified between men and women (0.91 D vs 0.21 D, respectively, p=0.84). However, both younger men and women were significantly more myopic than their older counterparts. The difference in degree of myopia was most marked when comparing those aged <60 years with people aged ≥60 years.

    View this table:
    Table 1

    Mean AL and ACD in men and women in the EPIC-Norfolk by age group

    Refractive data were available for both eyes in 2210 bilaterally phakic participants. High myopia (−6.00 D or less) accounted for 1.9%; moderate myopia (−5.99 D to −1.00 D), 17%; and low myopia (−0.99 D to −0.50 D), 8.1%. Emmetropia (−0.49 D to +0.49 D), low hypermetropia (+0.50 D to +0.99 D) and moderate hypermetropia (+1.00 D to +5.99 D) amounted to 21%, 13.7% and 37.9%, respectively, with high hypermetropia (>+6.00 D) accounting for 0.6%.

    Table 2 summarises a multivariate regression analysis that assessed the relationship between biometric covariates and age, sex, education, occupational class, height and weight. All putative predictive covariates were significantly, independently linked with AL, although occupational class was of borderline significance (p=0.06). This implies that, although greater body size was associated with greater AL of the eye, there were significant residual associations between sex, age and AL, although these were much attenuated after adjusting for height and weight. Of the six predictive characteristics examined, educational attainment (leaving school with at least one qualification vs no qualifications) seemed to exert the greatest magnitude of effect on AL. From the results of regression analysis shown in table 2, it can be seen that the effect of education (O level vs no O levels) on AL is 0.21 mm, which is equivalent to about the difference in axial length associated with an 8-cm difference in height, or about 25 years difference in age. In contrast, ACD did not appear to be significantly influenced by differences in education, occupation, height or weight. Significant independent relationships were identified between age, sex and ACD, with women and older people having shallower ACDs than men and younger people. Sex difference (0.058 mm) was approximately equivalent to that between people of 15 years age difference.

    View this table:
    Table 2

    Multivariate regression of AL and ACD with age and covariates in 2139 men and women in EPIC-Norfolk (β coefficient shows magnitude of change in dependent variable with change in the independent variable as shown)

    Discussion

    Refractive error in this predominantly white older UK population was associated with axial dimensions of the eye and sociodemographic characteristics, as we expected. Educational status was identified as the strongest determinant of axial length of the eye, with people who held higher educational qualifications having significantly longer eyes. This association persisted even after adjustment for the impact of height. Furthermore, older participants had smaller eyes and were, on average, more hypermetropic than the younger participants. This finding is somewhat counterintuitive in that it had been anticipated that age-related increases in lens nuclear sclerosis would drive the refraction towards myopia in older people. Indeed, this is almost certainly occurring, but in the context of the biometric and refractive data both showing higher rates of hypermetropia and shorter AL in older people, the impact of changes in lens refractive index is probably not the major determinant of refractive error in this cohort. Nonetheless, the fact that data on lens opacity for these people were not collected will probably have limited our ability to draw definitive conclusions on the factors responsible for refractive error in the cohort studied. Women were noted to have smaller eyes than men, both in terms of AL and ACD, but had no significant difference in refractive error.

    The data presented confirm that AL is associated with age, sex, education, occupational class, height and weight, and that these associations persist in a multivariate analysis, indicating that the influences that these factors exert are significant and independent from each other. The latter results add support to previously published findings from Singapore and Mongolia.17 19 In the current analysis, educational attainment appeared to be the most important environmental factor associated with AL, independent from effects of variation in age, sex, occupation, height and weight. The association between greater educational attainment and longer AL confirmed findings from Singapore where, in a multivariate analysis, the difference in AL associated with a 10-year difference in education was 0.60 mm greater for the more educated (p<0.001). In addition, after adjustment for age, sex, income, housing and education, non-manual workers were reported to have a 0.28-mm (p=0.02) longer AL than manual workers.17 Those Singaporeans in the highest pay bracket had an AL 0.45 mm longer than those on the lowest pay (>$3000/month vs ≤$1000/month, p=0.02). Hence, in Singapore too, it appeared that education, occupation and income were major factor influencing AL.17 However, the magnitude of the effect sizes for each of these factors was considerably greater among the Singaporeans, being three times greater for both education and occupation, although the relative contribution of each effect was very similar. It has been suggested previously that educational attainment is a marker for intensity of near work exposure in early life, and occupation similarly indicates the degree of near work exposure in later life.17 Data on Singaporean children have suggested that greater exposure to near work in childhood was associated with a higher incidence and more rapid progression of myopic refractive error.22 23 Our findings lend further weight to the belief that exposure to a lifestyle associated with greater near-work activity in early life drives refractive error towards myopia. The mechanism of myopic refractive error appears to be axial elongation.

    Interestingly, the sociodemographic factors influencing ACD appeared to differ from those associated with AL. Age and sex were significantly associated with ACD, but occupation, education, height and weight showed no identifiable relationship. It has previously been proposed that ACD is highly heritable (H2=70%)24 and that 66% of the risk of ACG (related to smaller axial size of the eye) is determined genetically. In the present study, women were found to have a shorter AL than men by 2.1% (0.51 mm), and similarly a shallower anterior chamber by 2.2% (0.07 mm). The sex difference in AL was identical to that found in the Los Angeles Latino Study (0.47 mm), although the difference in ACD was greater between Angelino men and women (0.12 mm).25 The ACD difference between English men and women was found to be almost identical to that identified by Tørnquist26 in normal Swedish individuals (0.09 mm). ACD is an important parameter because of its association with ACG. In a review by Lowe,27 mean ACD (including corneal thickness) in normal individuals was reported to lie between 3.03 and 3.15 mm, compared with 2.31 and 2.38 mm in ACG. Caucasians seen as part of the Baltimore Eye Study follow-up had mean ultrasound ACD of 3.07 mm (men) and 2.99 mm (women).28 The similarity between these mean figures for normal US Caucasians and our UK population is striking. Lowe27 further stated that the threshold for risk of angle-closure started at 2.5 mm, and increased below this level. In the EPIC-Norfolk cohort, 7.3% of women and 6.5% of men fell into this “at risk” category, values higher than expected.

    In summary, the present community-based study of an older, predominantly white UK population found the AL of the globe appeared to be strongly related to sociodemographic and economic indicators, but primarily to educational achievement. In contrast, ACD was found to vary with age and sex, but not with socioeconomic status. The findings of the current study lend further weight to the belief that ocular dimensions associated with refractive error and risk of glaucoma has strong environmental determinants.

    Acknowledgments

    EPIC was funded by the Medical Research Council, UK (G0401527), and Research into Ageing, UK (262). Mr Foster has received additional support from the Richard Desmond Charitable Trust (via Fight for Sight) and the Department for Health through the award made by the National Institute for Health Research to Moorfields Eye Hospital and the UCL Institute of Ophthalmology for a specialist Biomedical Research Centre for Ophthalmology. The views expressed in this article are those of the authors and not necessarily those of the Department for Health.

    References

    1. ↵
      1. Wensor MD,
      2. McCarty CA,
      3. Stanislavsky YL,
      4. et al
      . The prevalence of glaucoma in the Melbourne Visual Impairment Project. Ophthalmology 1998;105:733–9.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Attebo K,
      2. Ivers RQ,
      3. Mitchell P
      . Refractive errors in an older population: the Blue Mountains Eye Study. Ophthalmology 1999;106:1066–72.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Katz J,
      2. Tielsch JM,
      3. Sommer A
      . Prevalence and risk factors for refractive errors in an adult inner city population. Invest Ophthalmol Vis Sci 1997;38:334–40.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Wang Q,
      2. Klein BE,
      3. Klein R,
      4. et al
      . Refractive status in the Beaver Dam Eye Study. Invest Ophthalmol Vis Sci 1994;35:4344–7.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Kempen JH,
      2. Mitchell P,
      3. Lee KE,
      4. et al
      . The prevalence of refractive errors among adults in the United States, Western Europe and Australia. Arch Ophthalmol 2004;122:495–505.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Wong TY,
      2. Foster PJ,
      3. Hee J,
      4. et al
      . Prevalence and risk factors for refractive errors in adult Chinese in Singapore. Invest Ophthalmol Vis Sci 2000;41:2486–94.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Cheng CY,
      2. Hsu WM,
      3. Liu J-H,
      4. et al
      . Refractive errors in an elderly Chinese population in Taiwan: the Shihpai Eye Study. Invest Ophthalmol Vis Sci 2003;44:4630–8.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Lin LL,
      2. Shih YF,
      3. Hsiao CK,
      4. et al
      . Prevalence of myopia in Taiwanese schoolchildren: 1983 to 2000. Ann Acad Med Singap 2004;33:27–33.
      OpenUrlPubMed
    9. ↵
      1. Resnikoff S,
      2. Pascolini D,
      3. Etya'ale D,
      4. et al
      . Global data on visual impairment in the year 2002. Bull World Health Organ 2004;82:844–51.
      OpenUrlPubMedWeb of Science
    10. ↵
      1. Quigley HA,
      2. Broman AT
      . The number of persons with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90:262–7.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Knapp AA
      . Glaucoma in myopic eyes. Arch Ophthalmol 1926;55:35–7.
      OpenUrl
    12. ↵
      1. Möller HU
      . Excessive myopia and glaucoma. Acta Ophthalmol 1948;26:185–93.
      OpenUrlPubMed
    13. ↵
      1. Rosengren B
      . The etiology of acute glaucoma. Am J Ophthalmol 1953;36:488–92.
      OpenUrlPubMedWeb of Science
    14. ↵
      1. Alsbirk PH
      . Anterior chamber depth in Greenland Eskimos. I. A population study of variation with age and sex. Acta Ophthalmol 1974;52:551–64.
      OpenUrlPubMed
    15. ↵
      1. Foster PJ,
      2. Alsbirk PH,
      3. Baasanhu J,
      4. et al
      . Anterior chamber depth in Mongolians. Variation with age, sex and method of measurement. Am J Ophthalmol 1997;124:53–60.
      OpenUrlPubMedWeb of Science
    16. ↵
      1. Devereux JG,
      2. Foster PJ,
      3. Baasanhu J,
      4. et al
      . Anterior chamber depth measurement as a screening tool for primary angle-closure glaucoma in an East Asian population. Arch Ophthalmol 2000;118:257–63.
      OpenUrlPubMedWeb of Science
    17. ↵
      1. Wong TY,
      2. Foster PJ,
      3. Johnson GJ,
      4. et al
      . Education, socioeconomic status, and ocular dimensions in Chinese adults: the Tanjong Pagar Survey. Br J Ophthalmol 2002;86:963–8.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Wong TY,
      2. Foster PJ,
      3. Johnson GJ,
      4. et al
      . The relationship between ocular dimensions and refraction with adult stature: the Tanjong Pagar Survey. Invest Ophthalmol Vis Sci 2001;42:1237–42.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Wickremasinghe S,
      2. Foster PJ,
      3. Uranchimeg D,
      4. et al
      . Ocular biometry and refraction in Mongolian adults. Invest Ophthalmol Vis Sci 2004;45:776–83.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Riboli E
      . Nutrition and cancer: background and rationale of the European Prospective Investigation into Cancer and Nutrition (EPIC). Ann Oncol 1992;2:783–91.
      OpenUrl
    21. ↵
      1. Day NE,
      2. Oakes S,
      3. Lubin R,
      4. et al
      . EPIC-Norfolk: study design and characteristics of the cohort. European Prospective Investigation of Cancer. Br J Cancer 1999;80:95–103.
      OpenUrlPubMed
    22. ↵
      1. Saw SM,
      2. Chua WH,
      3. Hong CY,
      4. et al
      . Nearwork in early-onset myopia. Invest Ophthalmol Vis Sci 2002;43:332–9.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Saw SM,
      2. Tong L,
      3. Chua WH,
      4. et al
      . Incidence and progression of myopia in Singaporean school children. Invest Ophthalmol Vis Sci 2005;46:51–7.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Alsbirk PH
      . Anterior chamber depth and primary angle-closure glaucoma. II. A genetic study. Acta Ophthalmol 1975;53:436–49.
      OpenUrlCrossRefPubMed
    25. ↵
      1. Shufelt C,
      2. Fraser-Bell S,
      3. Ying-Lai M,
      4. et al
      . Refractive error, ocular biometry and lens opalescence in an adult population: The Los Angeles Latino Eye Study. Invest Ophthalmol Vis Sci 2005;46:4450–60.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Tørnquist R
      . Shallow anterior chamber in acute angle-closure. A clinical and genetic study. Acta Ophthalmol 1953;31:1–74.
      OpenUrlPubMed
    27. ↵
      1. Lowe RF
      . Primary angle-closure glaucoma: a review of ocular biometry. Aust J Ophthalmol 1977;5:9–17.
      OpenUrlCrossRef
    28. ↵
      1. Congdon NG,
      2. Qi Y,
      3. Quigley HA,
      4. et al
      . Biometry and primary angle-closure glaucoma among Chinese, white and black populations. Opthalmology 1997;104:1489–95.
      OpenUrlCrossRefPubMedWeb of Science
    View Abstract

    Footnotes

    • Funding Medical Research Council (London); Research into Ageing (London); The Richard Desmond Charitable Trust (via Fight for Sight, London); the NIHR BMRC UK (MEH & UCL-IOO).

    • Competing interests None.

    • Ethics approval This study was conducted with the approval of the East Norfolk and Waveney NHS Research Governance Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    Linked Articles

    • At a glance
      At a glance
      Harminder S Dua Arun D Singh
      British Journal of Ophthalmology 2010; 94 i-i Published Online First: 01 Jul 2010. doi: 10.1136/bjo.2010.189274