Distribution of the vertical cup to disc ratio in the right eye in the total population (n = 3228). The mean (SD) vertical cup to disc ratio was 0.40 (0.15).
Distribution of the vertical cup to disc ratio in the right eye in persons without glaucoma (n = 3081). The mean (SD) vertical cup to disc ratio was 0.39 (0.13).
Amerasinghe N, Wong TY, Wong W, Mitchell P, Shen SY, Loon S, Saw S, Foster PJ, Aung T, SiMES Study Group. Determinants of the Optic Cup to Disc Ratio in an Asian PopulationThe Singapore Malay Eye Study (SiMES). Arch Ophthalmol. 2008;126(8):1101-1108. doi:10.1001/archopht.126.8.1101
To describe the distribution and determinants of the optic cup to disc ratio (CDR) in Malay adults in Singapore.
This population-based, age-stratified study examined 3280 Malay people aged 40 to 80 years in Singapore. Participants underwent a standardized interview and an ocular examination. A slitlamp examination measured the vertical dimensions of the disc and cup, excluding areas of peripapillary atrophy and the Elschnig scleral ring.
Vertical CDR was recorded for 3228 right eyes and 3237 left eyes. The mean (SD) CDR was 0.40 (0.15) in both eyes. The CDR in the right eye increased with age (P < .001) and was greater in men vs women (age-adjusted CDR, 0.42 vs 0.39; P < .001). In multiple linear regression, significant determinants of greater CDR were increasing age, male sex, higher intraocular pressure (IOP), lower diastolic blood pressure, lower body mass index, and previous cataract surgery. Of these, higher IOP was the most important determinant of the CDR. After excluding 149 persons with glaucoma, male sex, higher IOP, lower diastolic blood pressure, lower body mass index, and diabetes mellitus were significant predictors of greater CDR.
Greater vertical CDR was related to male sex, higher IOP, lower diastolic blood pressure, and lower body mass index.
The morphologic characteristics of the optic disc are routinely assessed to screen, diagnose, and monitor disease in conditions such as glaucoma and optic neuropathies. Of various optic disc features, the vertical cup to disc ratio (CDR) is the most commonly used clinical measurement, particularly for the diagnosis of glaucoma.1
Systemic and ocular processes may affect the CDR, and understanding these factors may improve the clinical assessment of this sign. However, few population-based studies have assessed potential factors that may affect the CDR, and these have reported inconsistent results.1- 3The Blue Mountains Eye Study4and the Barbados Eye Study5found an increase in mean CDR with age. However, the Baltimore Eye Study1and the Rotterdam Study6did not. Regarding sex, the evidence again has been conflicting. Quigley et al7found that males had a larger CDR, but the Vellore Eye Study8did not find a significant difference. The Baltimore Eye Study1and the Rotterdam Study6found that the mean optic disc area was significantly larger in men than in women, but they did not find a statistically significant association with vertical CDR. A major limitation of these studies is that participants with and without glaucoma were often analyzed together. However, because the prevalence of glaucoma varies among populations, it is unclear how this might have affected study findings.
This study aims to describe the distribution and determinants of vertical CDR in a population-based cohort of Malay adults in Singapore. In particular, we examine associations in the whole population and then after excluding persons with glaucoma.
The Singapore Malay Eye Study was a population-based cross-sectional study of 3280 Malay residents of Singapore conducted between August 16, 2004, and July 10, 2006, and described previously.9In brief, an age-stratified random sampling procedure was used to select Malay people aged 40 to 80 years living in the southwestern part of Singapore.9Of 4168 persons eligible to participate, 3280 were examined (response rate, 78.7%). Nonparticipants were older but did not differ by sex or possession of a telephone in their homes (data not shown). This study was conducted in accordance with the World Medical Association Declaration of Helsinki. Ethics approval was obtained from the Singapore Eye Research Institute institutional review board.
Participants underwent a standardized interview, an ocular examination, and laboratory investigations at the Singapore Eye Research Institute Clinic. Slitlamp examination (Haag-Streit BQ 900; Haag-Streit AG, Koeniz, Switzerland) was performed before and after pupil dilation. Intraocular pressure (IOP) was measured using Goldmann applanation tonometry (Haag-Streit AG) and a standardized protocol.9Pupils were dilated with tropicamide, 1%, and phenylephrine hydrochloride, 2.5%.9
The optic disc was examined through a 78-diopter lens at ×10 magnification using the same technique as used in a population-based study of Chinese people in Singapore (Tanjong Pagar Survey).10,11The vertical dimensions of the disc and cup were measured using an eyepiece graticule, etched in 0.1 U. Measurements of vertical disc diameter excluded areas of peripapillary atrophy and the Elschnig scleral ring. The margins of the cup were defined by means of stereoscopic examination as the point of maximal inflexion of the contour. The vertical diameter of the cup was measured as the vertical distance between the points of maximal centrifugal extension of the cup between 11 to 1 o’clock and 5 to 7 o’clock. For small discs with no visible cup, the measurement was taken as the diameter of the emerging retinal vessel.10,11
Glaucoma was diagnosed using the International Society of Geographic and Epidemiological Ophthalmology classification.11Category 1 requires optic disc abnormality (vertical CDR or vertical CDR asymmetry ≥97.5th percentile or neuroretinal width between 11 to 1 o’clock and 5 to 7 o’clock <0.1 vertical CDR) and glaucomatous visual field defect. Category 2 requires a severely damaged optic disc (vertical CDR or vertical CDR asymmetry ≥99.5th percentile) in the absence of a satisfactory visual field test. Category 3 glaucoma is defined as blindness (corrected visual acuity <3/60), previous glaucoma surgery, or IOP greater than the 99.5th percentile if the optic discs cannot be examined.11
Height and weight were measured to calculate the body mass index (BMI), which is weight in kilograms divided by height in meters squared. Systolic and diastolic blood pressures (BPs) were measured with the participants seated after 5 minutes of rest using an automatic BP monitor and a standardized protocol.9Pulse pressurewas defined as systolic minus diastolic BP and ocular perfusion pressureas two-thirds of the mean arterial BP minus the IOP, where mean arterial BP is two-thirds of the diastolic plus one-third of the systolic value.
Axial length, anterior chamber depth (ACD), and corneal curvature in the horizontal and vertical meridians were measured using noncontact partial-coherence laser inferometry (IOLMaster; Carl Zeiss Meditec, Jena, Germany). Cataract was graded clinically using the Lens Opacities Classification System III.12Levels of nonfasting serum glucose, lipids (total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol), hemoglobin A1c, and creatinine were measured from venous blood collected from participants. Diabetes mellituswas defined as nonfasting glucose levels of 200 mg/dL or greater (to convert to millimoles per liter, multiply by 0.0555), self-report of diabetic medication use, or physician-diagnosed diabetes mellitus.
Statistical analysis was performed using a software program (SPSS version 11.5; SPSS Inc, Chicago, Illinois). Proportions were compared using the χ2test. Analysis of covariance models were initially used to estimate the mean vertical CDR, adjusted for age and sex. Multiple linear regression models were used to assess the independent association between CDR and risk factors. Final models are provided for right eyes only because the results for left eyes were largely similar. To test the normality and appropriateness of linear regression, we constructed Q-Q plots of vertical CDR that showed that vertical CDR was normally distributed between 0.1 and 0.7, which represented 99.2% (3203 of 3228 right eyes) of participants.
For linear regression, we provide the β coefficient and the partial R2. The β coefficient depends on the unit of measurement of the independent variable and cannot be directly compared with other independent variables in the model. The incremental partial R2allows assessment of the relative importance of the independent variables in the model. We conducted 2 subsidiary analyses. First, we repeated the analysis excluding glaucoma cases. Second, we repeated the analysis excluding patients with nuclear cataract (Lens Opacities Classification System III nuclear opacification or color ≥5.0).
The mean age of participants was 58.7 years. Vertical CDR was recorded for 3228 right eyes and 3237 left eyes. There were 149 patients with glaucoma, giving an overall prevalence of 4.5% in the population.
The mean (SD) vertical CDR was 0.40 (0.15) for both eyes. Figure 1shows the distribution of vertical CDR in the right eye in the entire study population, and Figure 2shows the distribution of vertical CDR in the right eye in persons without glaucoma. Table 1gives the distribution of vertical CDR by age group and sex for all persons (n = 3228) and after excluding patients with glaucoma (n = 3081). Vertical CDR increased with age in both groups (P < .001 and P = .04, respectively).
Table 2summarizes systemic predictors of vertical CDR in the right eye in the total population (n = 3228) and after excluding glaucoma cases (n = 3081). After controlling for age, the CDR was greater in men than in women (P < .001). Lower weight and BMI were associated with greater vertical CDR. Results were largely similar when patients with glaucoma were excluded (Table 2). In the analysis of eyes in which patients with nuclear cataract were excluded, vertical CDR remained greater in men than in women (0.41 vs 0.38; P < .001). Lower weight and BMI were still associated with greater vertical CDR (data not shown).
Ocular predictors of vertical CDR are summarized in Table 3. Vertical CDR was significantly larger with higher levels of IOP and lower levels of ocular perfusion pressure in the total population and in persons without glaucoma. Shallower ACD and previous cataract were associated with larger vertical CDR, but these associations were not significant after excluding patients with glaucoma (Table 3). Because ACD is associated with pseudophakia and aphakia, further multiple linear analyses were conducted using models containing only 1 of the 2 variables: ACD or previous cataract surgery. These analyses showed that ACD was not a significant determinant of vertical CDR in all persons and in those with glaucoma once cataract surgery was not in the model (data not shown). Cataract surgery, however, remained a significant determinant in all persons (P = .003) but not in models when persons with glaucoma were excluded (P = .16). In the final multiple linear regression, only cataract surgery was included in models for all persons.
Results were largely similar in analyses of eyes in which patients with nuclear cataract were excluded, in which higher levels of IOP and lower levels of ocular perfusion pressure were significantly associated with larger CDR (data not shown).
Table 4provides the final multiple linear regression models of vertical CDR in the total population and after excluding patients with glaucoma. In all persons, significant independent determinants of greater CDR were increasing age (P < .001), male sex (P < .001), lower diastolic BP (P = .002), lower BMI (P = .001), higher IOP (P < .001), and past cataract surgery (P < .004). Of these factors, IOP was the most important determinant of CDR, with the largest partial R2of 0.013. Factors in this model accounted for only 3.7% of all variation in CDR (adjusted R2 = 0.037).
When these data were analyzed excluding patients with glaucoma, male sex, diastolic BP, BMI, and IOP all remained significant determinants. However, age did not remain significantly associated with vertical CDR when persons with glaucoma were excluded. Also, diabetes mellitus was significantly associated with vertical CDR in persons without glaucoma. In this model, IOP was again the most important determinant of CDR, with the largest partial R2of 0.012. Factors in this model accounted for even less of the variation in CDR (adjusted R2 = 0.026).
In a subsidiary multiple regression analysis replacing diastolic BP with ocular perfusion pressure, significant independent determinants of greater CDR in all persons were increasing age, male sex, lower ocular perfusion pressure, lower BMI, higher IOP, and cataract surgery (data not shown). Of these factors, IOP was again the most important determinant of CDR, with the largest partial R2of 0.101. All factors accounted for only 3.5% of all variation in CDR (adjusted R2 = 0.035).
This population-based study in Asian Malays demonstrated that sex, IOP, BMI, diastolic BP, and ocular perfusion pressure were significant determinants of vertical CDR. The pattern of associations was largely similar after excluding persons with glaucoma and nuclear cataract, suggesting that these findings were not simply reflecting associations with glaucoma and were not affected by difficulties in assessing CDR in eyes with significant media opacity. Axial length was not associated with vertical CDR, and age was not a significant predictor of CDR once glaucoma was excluded.
The ocular and systemic factors studied herein accounted for less than 4% of the variation in vertical CDR, suggesting that other factors affect vertical CDR, including possible genetic factors. In support of this, studies13- 15suggest that the heritability of disc area and CDR is approximately 0.5.
In the present study, the most important determinant of vertical CDR was IOP, with the largest absolute value of partial R2. This was true even when patients with glaucoma were excluded. The importance of IOP is consistent with the Baltimore Eye Study1and the Blue Mountains Eye Study.2The Blue Mountains Eye Study2showed that for every 10–mm Hg increase in IOP, the CDR increased by 0.04. The Barbados Eye Study5also found that IOP was an important determinant in patients with open-angle glaucoma. The Beaver Dam Eye Study16assessed optic disc change during a 5-year interval and reported that change in IOP was significantly associated with change in vertical CDR.
Of the other ocular factors, past cataract surgery was associated with a greater vertical CDR, although this was not seen in persons without glaucoma. To our knowledge, no other study has examined this association in a population. The Barbados Eye Study5reported that cataract was associated with an increased risk of open-angle glaucoma, but the present study did not find a significant association between nuclear opacity and vertical CDR.
These data revealed other interesting associations. Lower BMI was associated with larger CDR, independent of IOP. The Barbados Eye Study5found that a higher BMI had some protective effect on the risk of open-angle glaucoma. Gasser et al17found that there was a tendency for patients with glaucoma to have a lower BMI than controls. However, underlying reasons for this association are unclear. We also found lower diastolic BP and ocular perfusion pressure to be associated with larger CDR, independent of IOP and even when patients with glaucoma were excluded. The Barbados Eye Study5also found associations of low diastolic BP and low BP to IOP ratio with risk of open-angle glaucoma. These observations provide evidence of possible mechanisms of reduced optic nerve head perfusion in glaucoma pathogenesis.
Two factors associated with CDR have been well studied. Age was a significant determinant of CDR only in the total sample, but not after excluding persons with glaucoma. This pattern is similar to that in the Rotterdam Study.6However, in the Blue Mountains Study,2each decade increase in age was associated with a 1.9% increase in mean CDR, and this relationship was still present after excluding patients with glaucoma or known optic disc disease. Garway-Heath et al4also reported that CDR increases by approximately 0.1 between ages 30 and 70 years. Despite conflicting results from different populations, histologic studies have reported a loss of ganglion cell nerve fibers with age.18- 22Jonas et al22estimated this loss to be approximately 0.36% per year, and Johnson et al19estimated it to be 0.625% per year. This may be an important area for further research.2Male sex was also a statistically significant determinant of larger CDR, consistent with other studies, reflecting possibly larger disc areas in men than in women.7
There are limitations of this study. First, the optic disc size of the participants was not measured, and optic disc size is known to affect CDR.23The Vellore Eye Study8found that high interindividual variability in optic disc and cup diameters results in variability in CDR. Therefore, CDR may have relatively low diagnostic power to differentiate between healthy eyes and those with early glaucoma. This power would increase significantly if the optic disc size is taken into account.23We collected Heidelberg Retina Tomograph II data on this population, and they are being analyzed to address these issues. Second, although the evaluation of CDR was quantified using an eyepiece graticule, it relied on subjective assessment of the disc and cup margins to perform the measurements. Training and agreement of CDR assessment with one of us (P.J.F.) using the Tanjong Pagar protocol was performed before the study commenced; however, the measurements were performed by multiple observers, and intergrader reproducibility was not assessed during the study. However, the effect of such measurement errors, likely to be random, is a dilution of strength of associations. Finally, myopic or tilted discs and media opacities could also have affected the optic disc evaluation and may have introduced unknown biases, although analyses excluding nuclear cataract had no effect on these associations.
In summary, in this population-based study, significant determinants of greater vertical CDR were male sex, higher IOP, lower diastolic BP, lower ocular perfusion pressure, lower BMI, past cataract surgery, and diabetes mellitus. Of these, higher IOP was the most important determinant of CDR. Nevertheless, the variables evaluated in this study accounted for less than 4% of the variation in vertical CDR, suggesting that other unknown factors may affect the CDR.
Correspondence:Tin Aung, FRCS, PhD, Glaucoma Department, Singapore National Eye Centre, 11 Third Hospital Ave, Singapore 168751 (email@example.com).
Submitted for Publication:May 29, 2007; final revision received December 22, 2007; accepted January 2, 2008.
SiMES Study Group:Tien Y. Wong, MBBS, MMED, MPH, FRCSE, FRANZCO, FAFPHM, PhD (principal investigator), Singapore Eye Research Institute, Singapore National Eye Centre, and National University of Singapore, Singapore, and University of Melbourne, East Melbourne, Victoria, Australia; Seang-Mei Saw, MBBS, MPH, PhD, FAMS (coprincipal investigator), National University of Singapore; Donald Tan, MBBS, FRCS, FRCOphth, FAMS, FRCS (coprincipal investigator), Singapore National Eye Centre, Singapore Eye Research Institute, National University of Singapore; Tin Aung, MBBS, MMED, FRCS, FRCOphth, PhD (coinvestigator), Singapore National Eye Centre, Singapore Eye Research Institute, National University of Singapore; Mohamad Rosman, MBBS, MMED (coinvestigator), Singapore National Eye Centre; Seng-Chee Loon, MBBS, MMED, FAMS (coinvestigator), Singapore National Eye Centre; Jing Liang Loo, MBBS, MRCS Ed, MMED (coinvestigator), Singapore National Eye Centre; Sunny Shen Yu, MBBS, MMED (coinvestigator), Singapore National Eye Centre; Tai E. Shyong, MB, ChB, FRCP(UK) (coinvestigator), Singapore General Hospital, National University of Singapore–Genome Institute of Singapore Centre for Molecular Epidemiology, National University of Singapore; Ronald Klein, MD, MPH (external advisory board member), University of Wisconsin, Madison; Barbara Klein, MD, MPH (external advisory board member), University of Wisconsin, Madison; James Tielsch, BS, MHS, PhD (external advisory board member), Johns Hopkins University, Baltimore, Maryland; Paul Mitchell, MBBS(Hons), MD, FRACO, FRACS, FRCOphth, FAFPHM, PhD (external advisory board member), University of Sydney, Sydney, Australia; Paul Foster, BMedSci, BMBS, PhD, FRCS(Edin) (external advisory board member), Institute of Ophthalmology and Moorfields Hospital, London, England; Jie Jin Wang, MBBS (equivalent), MMED (Clin Epi), PhD, MApplStat (external advisory board member), University of Sydney.
Financial Disclosure:None reported.
Funding/Support:This study was supported by grant 0796/2003 from the National Medical Research Council and by grant 501/1/25-5 from the Biomedical Research Council, with support from the Singapore Prospective Study Program and the Singapore Tissue Network, A*STAR.