Abstract
Purpose:
To compare peripapillary retinal nerve fiber layer (RNFL) thickness among healthy adults by race/ethnicity and identify determinants of RNFL thickness.
Design:
Population-based, cross-sectional study
Participants:
Data from 6,133 individuals (11,585 eyes) from three population-based cohort studies in Los Angeles county, CA who were ≥ 50 years of age with self-described African, Chinese, or Latin American ancestry
Methods:
We measured RNFL thickness and optic nerve head (ONH) parameters using the Cirrus HD-OCT 4000. Multivariable linear mixed regression analysis was used to evaluate factors associated with RNFL thickness, after accounting for relatedness, among participants without ocular diseases.
Main Outcome Measures:
Determinants and modifiers of RNFL thickness
Results:
The mean age of the study participants was 60.1 (SD:7.4) years. Among the 3 groups, African Americans had the lowest RNFL thickness and smallest cup-to-disc ratio (CDR) across all quadrants, and Chinese Americans had the largest CDR and largest disc area after adjusting for age and sex (all P < 0.05). Per each 10-year older age group, the average RNFL thickness was 2.5 (95% CI: 1.8-3.1), 2.8 (95% CI: 2.3-3.3), and 3.5 (95% CI: 2.9-4.1) μm thinner for African, Chinese, and Latin Americans, respectively (age trend P < 0.05 and interaction P = 0.041). In the multivariable model, African Americans compared to Chinese Americans, older age, male sex, hypertension, diabetes, greater axial length (AL), bigger disc area, and lower scan signal strength were associated with thinner average RNFL. Race/ethnicity, age, AL, disc area, and scan signal strength were consistently associated with RNFL thickness in all quadrants, while sex, hypertension, and diabetes were associated with RNFL thickness in select quadrants. Age and race/ethnicity explained the greatest proportion of variance of RNFL thickness.
Conclusions:
Clinically important differences in RNFL thickness are present in healthy adults ≥ 50 years of age from different racial/ethnic groups of the same age, with the thinnest measures observed in African Americans. This race/ethnicity difference remains after accounting for disc size and AL. Further, age-related RNFL thinning differs by race/ethnicity. Longitudinal studies are needed to verify our findings and assess the influence of race/ethnicity in the clinical application of RNFL thickness.
Changes in the ONH and thinning of the RNFL occur with aging and optic neuropathies, including glaucoma.1-3 Optical coherence tomography (OCT) has allowed high-resolution visualization and quantification of the structure of the optic nerve head (ONH) and the peripapillary retinal nerve fiber layer (RNFL)4 and early detection of many optic neuropathies. Currently, commercial OCT devices use their own built-in normative database as the reference for detecting RNFL abnormalities.
However, the clinical application and interpretation of OCT-based RNFL measurements may be compromised in minority populations, who are underrepresented in the normative reference databases. Hispanics and African Americans, who are at high risk for developing glaucoma, are underrepresented in many OCT normative databases, which may consist of 200 to 500 participants5 overall. For example, the normative OCT study by Knight et al included 35 Hispanics, representing 12% of the sample.2,6 Existing studies of OCT that have included minorities suggest that Hispanics and Asians have greater RNFL thickness overall compared to non-Hispanic whites, while the findings for African Americans compared to other race/ethnicities have been mixed.7,8 Reasons for the inconsistencies may include lack of quadrant-specific reporting and low statistical power due to small numbers of participants. Further, there are well-known racial/ethnic differences in ocular characteristics that influence OCT-based measurement of RNFL thickness, such as optic disc area9-12 and refractive error13. As a result, there is an important need for a better understanding of racial/ethnic differences in OCT-based measurement of RNFL thickness.14
The aims of this study were to provide a robust assessment of 1) the magnitude of racial/ethnic differences in RNFL thickness and neuroretinal rim in understudied, older minority Americans overall and by age group, and 2) how these differences are modified by ocular factors and health conditions, such as axial length (AL), BMI, and diabetes. We created a large database of optic disc scans (N=11,585 healthy eyes from 6,133 individuals) obtained through Cirrus HD-OCT 4000 from three population-based, similarly designed cohort studies of minority Americans: the Los Angeles Latino Eye Study (LALES), the Chinese American Eye Study (CHES), and the African American Eye Disease Study (AFEDS).
MATERIAL AND METHODS
Study design and participants
We pooled epidemiological, ocular, and OCT imaging data from three population-based studies of eye diseases conducted in Los Angeles, California: the 8-year follow up study of LALES (LALES III) conducted between 2009 and 2014, CHES conducted between 2009 and 2013, and AFEDS conducted between 2014 and 2018. The three studies were designed by the same investigative team and shared identical study protocols. The methods for each study have been described previously.15-17 Briefly, participants were eligible for each respective study depending on self-reported race/ethnicity, being 40 years and older for LALES baseline and AFEDS and 50 years and older for CHES, living within respective recruitment communities. Therefore, this current investigation was restricted to participants 50 years and older to ensure comparability across studies. Trained interviewers completed standardized in-home interviews, and participants were then invited to complete comprehensive clinical examinations by ophthalmic technicians and ophthalmologists. OCT scans of the three studies were completed between 2010 and 2016. Written informed consent was obtained from all participants. Approvals were obtained for all studies from the Los Angeles County/University of Southern California Medical Center Institutional Review Board and adhered to the tenets of the Declaration of Helsinki.
Demographic and clinical data
All studies conducted standardized in-person interviews and comprehensive ocular examinations, which collected demographic, clinic, and ophthalmic information, including gender, age, height, weight, waist-hip ratio, blood lipids, hypertension, being diabetic, and AL. Two consecutive measurements of systolic and diastolic blood pressure were obtained using the random zero sphygmomanometer and participants were categorized as hypertensive with a systolic blood pressure >140mmHg or diastolic blood pressure > 90 mmHg. Non-fasting blood was utilized for lipid measurements and measured using the Cholestech LDX System (Alene, Waltham, MA). Glycosylated hemoglobin was measured using the DCA 2000+ System (Bayer Corporation, Tarrytown, NY). Participants were considered to have diabetes if 1) the participant reported a history of diabetes and was being treated with medication or insulin or 2) HbA1c measured at 6.5% or higher. AL was measured using an ultrasonic A-scan/pachymeter DGH 4000B SBH IOL Computation module (DGH Tech Inc., Exton, PA).
Optical coherence tomography imaging
OCT imaging of the RNFL and ONH was performed using the Cirrus HD-OCT 4000 (Carl Zeiss Meditec, Dublin, CA, USA). In the AFEDS population, the Cirrus HD-OCT 4000 was used in the study until it was upgraded to the Zeiss Cirrus HD-OCT 5000 with AngioPlex OCT in April 2016. There were 3,318 AFEDS participants that were recruited after April 2016 imaged on the Cirrus HD-OCT 5000 and therefore were not approached for Cirrus HD-OCT 4000 imaging. Compared to the participants approached April 2016 and later, those participants with Cirrus HD-OCT 4000 scans were younger and shorter AL (Supplementary Table 1, available at http://www.aaojournal.org). Trained ophthalmic technicians collected Optic Disc Cube 200x200 scans for the right (OD) and left (OS) eye separately. The Cirrus algorithm identifies the center of the optic disc and automatically places a 360° view with a 3.46 mm diameter. We utilized the Zeiss software with the default parameters to perform segmentation and calculated the RNFL thickness. First, the anterior and posterior boundaries of the RNFL are delineated. The system then calculates the RNFL thickness by counting the number of pixels between the anterior and posterior boundaries along each point on the circle of the A-scan. RNFL thickness measurements are categorized into clock hour, quadrants, and by aggregate average. To assess ONH measures, the algorithm identifies the termination of Bruch’s membrane to determine the disc edge and then extracts the rim width around the optic disc by measuring the thickness of the neuroretinal tissue in the optic nerve. We then extracted the ONH parameters, including disc area, rim area, cup-to-disc area ratio (CDR), vertical cup-to-disc ratio (VCDR), and cup volume. The CDR is calculated using the square root of the ratio of the cup area to the disc area. The VCDR is the ratio of the cup diameter to the disc diameter in a vertical meridian through the cup center. The cup volume is the 3-dimensional measurement defined as the volume between a plane created at 200 μm offset to the plane of the disc and the vitreoretinal interface.2
Inclusion and exclusion criteria
As recommended by the manufacturer for high quality, we retained only scans with a signal strength ≥7 out of 10 for our analysis.18,19 To make inferences on a normal population, we also excluded participants that had a diagnosis of glaucoma, diabetic retinopathy, macular degeneration, and cataract related phenotypes, determined at the clinic examination of each study. A glaucoma diagnosis was determined by the comprehensive ophthalmologist which included evaluation for an optic nerve rim defect characteristic of glaucoma and visual field assessment. Additionally, we excluded participants with any reported history of glaucoma or cataract procedures, including glaucoma drain tube, glaucoma laser surgery, and cataract extraction. Participants with extreme refractive error (spherical equivalent <-6D and >6D) were also excluded. We excluded scans for any potential segmentation errors, i.e. scans that have RNFL thickness values equal to zero in any clock hour. We also assessed symmetry across eyes and excluded participants if the difference in average or any quadrant-specific RNFL thickness between eyes was greater than 30 μm as suggested by Budenz and Mwanza et al.20,21 If participants had multiple high quality scans that passed these quality control measures, only one scan per eye was randomly selected.
Statistical analyses
We considered demographic and biological variables including age, sex, body mass index ([BMI] normal, <25 kg/m2; overweight, 25 - <30 kg/m2; obese, ≥30 kg/m2), waist-hip ratio, ever smoker, high density lipoprotein, low density lipoprotein, hypertension (Yes vs. No), and diabetic status (Yes vs. No) as candidates for our multivariable models. These variables have been shown or suspected to be associated with RNFL thickness in prior studies and were parameterized similarly.22-25 We also considered ocular variables including AL, intraocular pressure (IOP), refractive error, disc area, and signal strength. Counts, proportions, means, and standard deviations were used to summarize the demographic and clinical characteristics.
We used linear mixed models (LMM) to assess the relationship between continuous RNFL thickness measures and demographic and clinical variables while allowing for the relatedness between eyes by including a random effect in the model. Analyses adjusting for age, race/ethnicity, and gender were performed first and all variables that were significant at P < 0.1 in this analysis were considered in the final multivariable model. Backward stepwise regression was performed to remove variables with P > 0.05. Estimated marginal means with 95% confidence intervals (CIs) were calculated to compare the differences of ONH and RNFL thickness across different race/ethnicity and demographic characteristics after adjusting for other covariates.
Quantile regression was used to compare the effects of race/ethnicity on RNFL thickness at the fifth percentile. Additionally, we calculated square semipartial correlation coefficients (SR2) to estimate the unique contribution of each covariate to the variation in outcomes. We plotted the average RNFL thickness at the predicted values and the fifth percentile by race/ethnicity to describe “extreme” values adjusting for age and sex, for better comparisons with results from previous OCT normative studies.7
Sensitivity analyses were performed using two subsets of the final dataset including analyzing 1) higher quality scans using a signal strength ≥9, and 2) randomly selecting one eye per participant and running regression models without a random intercept. Furthermore, we evaluated the role of glaucoma in our multiethnic population by comparing the RNFL distributions in participants with glaucoma to those in health participants. There were 521 participants with a clinical diagnosis of glaucoma or glaucoma suspect based on evidence of glaucomatous visual field abnormality and/or glaucomatous optic disc damage in at least one eye, who were initially excluded from the primary analyses. Statistical analyses were performed using R (Version 3.5.2).
RESULTS
Across the three cohorts, there were 10,724 participants that were 50 years and older (Figure 1). Among them, 3,249 were excluded because they had ocular diseases, a history of glaucoma or cataract procedures, or extreme refractive error. Among the remaining healthy individuals who were free of ocular diseases, 6703 (89.7%) had OCT scans and 6,133 (82.0 %) had high-quality scans that were used for our primary analyses (Table 1). Participants that were excluded based on signal strength and other scan quality measures were older than participants with better scans with a mean age of 66.02 to 60.1 (P < 0.0001).
Figure 1.
Flow chart of the Study Cohort. Abbreviations: AFEDS, African American Eye Disease Study; CHES, Chinese American Eye Disease Study; LALES, Los Angeles Latino Eye Study
Table 1.
Characteristics of 50+ years old Participants with Healthy Eyes in the Combined dataset.
| Participants with Healthy Eyes | |||
|---|---|---|---|
| Variable | Total (n=7475) |
Those with high-quality OCT scans (primary analysis cohort) (n=6133) |
P for participants with vs. without high-quality OCT |
| n (%) | n (%) | P | |
| Female | 4771 (63.83) | 3923 (63.97) | 0.880 |
| Age | < 0.001 | ||
| 50-59 years | 3840 (51.37) | 3271 (53.33) | |
| 60-69 years | 2598 (34.76) | 2132 (34.76) | |
| 70-79 years | 857 (11.46) | 637 (10.39) | |
| 80+ years | 180 (2.41) | 93 (1.52) | |
| Race/ethnicity | 0.007 | ||
| African Americans | 1607 (21.5) | 1311 (21.38) | |
| Chinese Americans | 3281 (43.89) | 2843 (46.36) | |
| Latino Americans | 2587 (34.61) | 1979 (32.27) | |
| BMI (kg/m2) | |||
| Normal (<25) | 2497 (34.19) | 2161 (35.4) | 0.334 |
| Overweight (25 - <30) | 2533 (34.68) | 2087 (34.19) | |
| Obese (≥30) | 2274 (31.13) | 1857 (30.42) | |
| Diabetic | 1385 (19.53) | 1069 (18.28) | 0.075 |
| Ever Smokers | 2120 (29.28) | 1723 (28.64) | 0.428 |
| Hypertensiveb | 2140 (28.91) | 1719 (28.12) | 0.320 |
| Variable | Mean (SD) | Mean (SD) | |
| Waist-Hip Ratio | 0.88 (0.08) | 0.87 (0.08) | 0.297 |
| HDL (mg/dL) | 50.67 (16.77) | 50.55 (16.75) | 0.705 |
| LDL (mg/dL) | 104.95 (33.14) | 104.91 (33.12) | 0.950 |
| Axial Length (mm) | 23.56 (1.25) | 23.52 (1.17) | 0.047 |
| IOP (mmHg) | 14.81 (2.73) | 14.79 (2.64) | 0.645 |
Abbreviations: HbA1c, Hemoglobin A1c; HDL, high density lipoprotein, LDL, low density lipoprotein; IOP, intraocular pressure.
Univariate P-value
Hypertension with systolic blood pressure (SBP) >140mmHg or diastolic blood pressure (DBP) >90 mmHg
In our analysis cohort (i.e. healthy participants with high quality OCT data), there were 2,843 (46.4%) Chinese Americans, 1,979 (32.3%) Latinos and 1,311 (21.4%) African Americans (Table 1). The mean age of the study population was 60.1 (SD=7.4). As expected, there were more women in this study (n=3,923, 64.0%) than men. Complete data on clinical, demographic, and ocular variables were available for 4,615 healthy participants for LMM multivariable analyses of their associations with RNFL thickness.
We observed racial/ethnic differences across different RNFL thickness and relevant ONH measures, and the differences remained after adjusting for age and gender (Table 2). On average, African Americans had thinner RNFL than their Chinese American and Latino counterparts (all P <0.05 using a Tukey test) and the magnitude of racial/ethnic difference was greatest in the temporal quadrant (mean difference= −13.47 and −4.18 μm, comparing African Americans to Chinese Americans and Latinos, respectively). Latino Americans had thicker nasal RNFL and thinner temporal RNFL than Chinese Americans (pairwise P < 0.05) but had similar superior and inferior RNFL thickness. We estimated the mean differences between the temporal and nasal quadrants by race/ethnicity; we identified that in Chinese Americans, temporal quadrants were thicker than their nasal quadrants opposite what was observed in Hispanic whites and African Americans (all P < 0.001). The associations with race/ethnicity remained significant in our multivariable analyses (see Tables 3 and 4 below). In addition, compared with Chinese Americans, African Americans had greater cup volume but smaller disc area and lower average CDR (all pairwise P <0.05); however, the magnitude of these differences was relatively small (Table 2). Similar race/ethnicity differences were observed when we restricted our analyses to OCT scans with a signal strength ≥9 or one eye per person (Supplementary Table 2, available at http://www.aaojournal.org).
Table 2.
Comparison of Optic Nerve Head Parameters and Retinal Nerve Fiber Layer Thickness Measurements Across Race/ Ethnicities.
| Variable | Race/Ethnicities Combined Mean (SD) |
African Americans Mean (SD) |
Chinese Americans Mean (SD) |
Latinos Mean (SD) |
Pairwise differences by Race/Ethnicitya |
|---|---|---|---|---|---|
| Optic Nerve Head | |||||
| Rim Area (mm2)b | 1.32 (0.25) | 1.33 (0.26) | 1.33 (0.24) | 1.31 (0.25) | A>L |
| Disc Area (mm2) | 2.02 (0.39) | 1.98 (0.38) | 2.06 (0.4) | 1.98 (0.36) | C>A, C>L |
| Average CDR | 0.55 (0.14) | 0.53 (0.15) | 0.56 (0.14) | 0.55 (0.14) | C>A, C>L, L>A |
| VCDR | 0.51 (0.13) | 0.5 (0.14) | 0.51 (0.13) | 0.51 (0.13) | NS |
| Cup Volume (mm3) | 0.18 (0.15) | 0.19 (0.16) | 0.17 (0.15) | 0.18 (0.14) | A>C |
| RNFL | |||||
| Average RNFL (μm) | 95.11 (10.11) | 90.87 (10.38) | 97.11 (9.28) | 95.09 (10.17) | C>A, C>L, L>A |
| Temporal RNFL (μm) | 65.56 (12.16) | 58.02 (9.81) | 71.49 (11.69) | 62.2 (10) | C>A, C>L, L>A |
| Superior RNFL (μm) | 118.58 (16.48) | 114.21 (16.83) | 120.3 (16.36) | 119.03 (15.89) | C>A, L>A |
| Nasal RNFL (μm) | 71.24 (11.09) | 71.56 (11.31) | 69.9 (10.42) | 72.91 (11.6) | A>C, L>C, L>A |
| Inferior RNFL (μm) | 125.05 (17.3) | 119.64 (17.63) | 126.74 (16.74) | 126.22 (17.13) | C>A, L>A |
Abbreviations: CDR, Cup-disc ratio; VCDR, vertical cup-disc ratio;
Significant differences between each ethnicity group and the direction (A, African Americans; C, Chinese Americans; L, Latinos) at Tukey’s P<0.05. P-values were estimated using a linear mixed model after adjusting for age and sex.
The pairwise Chinese American versus Latino comparison was marginally significant after adjusting for age and sex (P=0.085).
Table 3.
Linear Mixed Regression Results of Average RNFL Thickness.
| Variable | Average RNFL thickness Model 1a |
Average RNFL thickness Model 2b |
||
|---|---|---|---|---|
| β (SE) | P | β (SE) | P | |
| Race/Ethnicity | ||||
| Chinese American | Reference | Reference | ||
| Latinos | −1.45 (0.33) | <0.001 | −1.73 (0.32) | <0.001 |
| African Americans | −5.55 (0.35) | <0.001 | −5.39 (0.33) | <0.001 |
| Age | ||||
| 50-59 years | Reference | Reference | ||
| 60-69 years | −2.8 (0.3) | <0.001 | −2.05 (0.28) | <0.001 |
| 70-79 years | −5.84 (0.48) | <0.001 | −4.91 (0.45) | <0.001 |
| 80+ years | −7.16 (1.24) | <0.001 | −5.94 (1.14) | <0.001 |
| Sex (Female vs Male) | 1.15 (0.29) | <0.001 | 0.73 (0.27) | 0.01 |
| BMI (kg/m2) | ||||
| Normal (<25) | Reference | |||
| Overweight (25 - <30) | 0.22 (0.35) | 0.53 | ||
| Obese (≥30) | −0.09 (0.42) | 0.83 | ||
| Waist-Hip Ratio (per 1 unit) | −4.41 (2.01) | 0.03 | ||
| Ever Smokers (Yes vs. No) | 0.1 (0.34) | 0.77 | ||
| HDL (per 1 mg/dL | 0.01 (0.01) | 0.68 | ||
| LDL (per 1 mg/dL) | 0.01 (0.01) | 0.09 | ||
| Hypertension (Yes vs. No) | −0.71 (0.32) | 0.03 | −0.62 (0.3) | 0.04 |
| Diabetes (Yes vs. No) | −0.95 (0.37) | 0.01 | −0.85 (0.35) | 0.02 |
| Axial Length (per 1 mm) | −1.24 (0.09) | <0.001 | −0.96 (0.09) | <0.001 |
| IOP (per 1 mmHg) | −0.08 (0.04) | 0.06 | ||
| Disc Area (per 1 mm2) | 4.54 (0.24) | <0.001 | 4.01 (0.23) | <0.001 |
| Signal strength of scan | 1.18 (0.07) | <0.001 | 1.06 (0.07) | <0.001 |
Abbreviations: HbA1c, Hemoglobin A1c; HDL, high density lipoprotein, LDL, low density lipoprotein; IOP, intraocular pressure
Model 1- Estimates for age, gender, and ethnicity were estimated from a model that included these three variables only. Effects for other variables were estimated with additional adjustment for age, gender, and ethnicity.
Model 2- Effects were estimated from a model that included all variables discovered in Model 1with P < 0.1 and used a backward stepwise exclusion of all variables with P > 0.05.
Table 4.
The Race/Ethnicity Effect on Average RNFL Thickness Estimated with Different Adjustment for Potential Confounders.
| Regression Coefficient β (SE) | ||
|---|---|---|
| Confounders considered | Latinos vs Chinese Americans |
African Americans vs Chinese Americans |
| - | −2.09 (0.32) | −5.97 (0.34) |
| Age, Sex | −1.75 (0.32) | −5.54 (0.34) |
| Age, Sex, Axial Length | −2.41 (0.32) | −5.77 (0.33) |
| Age, Sex, Disc Area | −1.34 (0.31) | −5.27 (0.33) |
| Age, Sex, Axial Length, Disc Area | −1.93 (0.31) | −5.49 (0.33) |
| Age, Sex, Axial Length, Disc Area, and other covariates identifieda | −1.78 (0.32) | −5.39 (0.33) |
All covariates include all variables discovered in the multivariable regression model including hypertension, diabetes status, and disc signal strength.
Figure 2 includes both the average and the fifth percentile RNFL thickness by clock-hour for each race/ethnicity after adjusting for age and gender. Across race/ethnicity groups, similar differences in RNFL thickness were present when considering the fifth percentile (group level quantile regression P < 0.05; Supplementary Table 3, available at http://www.aaojournal.org). Additionally, the RNFL thickness in those with a clinical diagnosis of glaucoma is thinner compared to those with healthy eyes across all clock hours in each race/ethnicity (Supplementary Figure 1, available at http://www.aaojournal.org). The participants with a glaucoma diagnosis were similar to the RNFL thickness in healthy eyes at the fifth percentile adjusting for age.
Figure 2.
Line series plot of RNFL thickness by clock-hour stratified by race/ethnicity. Statistically significant differences adjusted for age and gender at clock hour using estimated marginal between studies with adjustment using Tukey’s test. Abbreviations: C-A, Significant difference between Chinese Americans and African Americans (Tukey’s P < 0.05); C-L, Significant difference between Chinese Americans and African Americans (Tukey’s P < 0.05); C-L, Significant difference between Chinese Americans and Latinos (Tukey’s P < 0.05); A-L, Significant difference between African Americans and Latinos (Tukey’s P < 0.05)
Both average and quadrant specific RNFL thickness were lower in older ages (Figure 3) across all three racial/ethnic groups (P for age trend<0.05), respectively. However, the estimated age-related difference in average RNFL thickness, especially RNFL thickness in nasal quadrant, was greater in Latinos than the other two groups. Per each 10-year increase in age group, average RNFL thickness was 2.5, 2.8, and 3.5 μm thinner for African, Chinese, and Latin American adults respectively (P for age-race interaction=0.041). For quadrant-specific RNFL, age-related RNFL loss in the entire cohort was more pronounced in the inferior and superior quadrants (4.8 and 4.6 μm thinner per each 10-year increase in age, respectively) than in the temporal and nasal quadrants (1.1 and 1.3 μm, respectively). Figure 4 summarizes the difference in RNFL thickness per each 10-year increase in age across clock hours by race/ethnicity group. The greatest age-related difference in RNFL thickness was seen in the clock hour seven region of Latinos (mean difference in RNFL thickness= −5.93 μm per 10 years older in age).
Figure 3.
Linear mixed regression line plots of A) Average and B) Quadrant Specific RNFL thickness with age across three race/ethnicities
Figure 4.
Race/ethnicity-specific differences in RNFL thickness for each clock-hour per 10-year increase in age
Table 3 summarizes the LMM regression of average RNFL thickness on potentially associated factors. The final multivariable model for average RNFL thickness included age, sex, race/ethnicity, hypertension, being diabetic, AL, disc area, and signal strength. Similar associations were observed for quadrant specific RNFL thickness across all quadrants (Supplementary Table 4, available at http://www.aaojournal.org). In total, these factors explained between 7% to 25% of the variance in quadrant specific RNFL thickness and 17% of the variance in average RNFL thickness (Supplementary Table 5, available at http://www.aaojournal.org). Race/ethnicity, age, disc area, and AL explained the greatest proportion of the variances across RNFL thickness quadrants.
We also assessed potential contributors to the observed racial/ethnic difference in average RNFL thickness (Table 4). Even though AL and disc area were found to be potential confounders, the race/ethnicity difference in average RNFL thickness remained after adjustment for AL, disc area, and other covariates identified in the multivariable regression.
DISCUSSION
The present pooled analysis is the largest population-based study that evaluated the normative values and determinants of OCT-measured RNFL thickness in minority Americans. This study is comprised of data from the AFEDS, CHES, and LALES, three of the largest population-based studies of age-related eye diseases in African, Chinese, and Latin Americans, respectively. In our population of those ≥50 years old, we observed notable differences in RNFL thickness among healthy eyes by race/ethnicity with the thinnest measures generally found among African Americans, followed by Latinos and Chinese Americans. These differences remained after adjusting for potential confounding factors such as disc area and AL.
Even though OCT has been widely used in detecting early glaucoma, there are limited population-based data on OCT-measured RNFL thickness in populations at higher risk for glaucoma, such as African Americans and Latinos. While previous findings have suggested that there are racial differences in retinal structure, normative databases do not typically account for race6. The recent E3 Consortium meta-analysis identified a RNFL thinning of 3.5 μm per year in a cohort of predominantly white Europeans.25 The normative population used to construct the original Zeiss Cirrus OCT database consisted of 284 participants of European, Chinese, African, and Hispanic descent;2 however, the plurality of these participants (n=122) were of European descent. Considering the scarcity of representative data on retinal structure and specifically RNFL thickness in minority Americans, population-based studies including LALES, CHES, and AFEDS were created and OCT-based measurements were collected systematically.
Among our AFEDS and LALES populations, we observed attenuated average RNFL thickness values for our African and Latino Americans as compared to findings by Girkin et al.8 However, their higher results may be explained by their inclusion of younger participants (>18 years old) and differences in modeling techniques (full covariates not included in models). Additionally, our findings of Chinese Americans are consistent with those from the Singapore Epidemiology of Eye Disease (SEED) study, where the average and fifth percentile values were reported to be 95.7 and 80.0 μm, respectively.24 Similarly with the SEED study, Chinese participants had the thinnest nasal quadrants while other ethnic groups had the thinnest temporal quadrants.24 With SEED identified similar covariates associated with RNFL thickness including AL, disc area, hypertension, and diabetes for Chinese26. Our findings of Asian and Hispanic participants having greater RNFL thickness than African American participants are consistent with those from previous studies.2,7,27 Our study did not include data from participants of European descent for direct comparison. In general, previous studies that included European descendants reported that the RNFL thickness in these populations are the thinnest compared to all their racial/ethnic counterparts including African descendants.2,7,10 However, these values can vary quite markedly as shown in the recent E3 Consortium with average RNFL thickness ranging from 86.8μm to 104.7 μm across multiple European populations.25 This highlights the need to control for age, imaging parameters, or variation in the optic nerve head beyond the comparisons between race.
Other important differential patterns emerged across our 3 racial/ethnic groups. First, even among our youngest participants (50 years), RNFL thickness measures in temporal, superior, and inferior quadrants were approximately 4-13 μm thinner in African Americans than Chinese or Latino Americans, and this difference is equivalent to 15+ years of age-related RNFL thickness loss. Second, while Latino Americans have thicker RNFL on average than African Americans at younger ages (e.g. 50s), they appear to have a greater rate of age-related RNFL loss and therefore have similar average RNFL thickness to that of African Americans in older ages (>70). We examined a subset of 50-year-old and 70-year-old participants and show significant differences at age 50 between Latinos and African Americans (mean difference = 4.46, P < 0.0001) but less so at age 70 (mean difference = 2.03, P=0.05). These observations are consistent with the racial/ethnic differences observed for age-specific prevalence of open-angle glaucoma.28,29 Not only do African Americans have 2-4 times higher risk for open-angle glaucoma than non-Hispanic whites, Latinos, and Asians, the disease usually impacts African Americans at an earlier age than other racial/ethnic groups.30-32 In Latino Americans, the risk of open-angle glaucoma has been shown to rise rapidly after age 65. It is possible that the higher risk of glaucoma in African Americans seen in much younger ages may be, in part, due to the lower nerve tissue reserve, whereas the higher risk for glaucoma seen in Latin Americans at older ages may be due to more rapid age-related RNFL thinning.
The racial/ethnic differences in OCT-measured RNFL thickness are in part, but not completely, resulting from racial/ethnic differences in ocular characteristics, especially disc area and AL. Because OCT instruments fix the scan using a 3.4 mm diameter around the ONH regardless of the disc area, the measure of RNFL thickness samples an area with more optic nerve fibers among individuals with larger optic discs.33 Consistently, in the current study, disc area was positively correlated with RNFL thickness in all three racial/ethnic groups, and the differences of RNFL thickness by race/ethnicity were indeed attenuated after controlling for differences in disc area. However, contrary to findings from previous studies that individuals with African descent have the largest disc area,2,8,9,11,12,34,35 we found only small, albeit significant, difference in disc area across the three racial/ethnic groups, with Chinese Americans having larger disc area than African Americans and Latinos. This association remained even after adjustment for AL (results not shown). This inconsistency in race/ethnicity differences in disc area between studies may be due to differences in participant recruitment (e.g. population-based vs. clinic-based) or differences in how disc area was ascertained (e.g. OCT vs. fundus). This was also reflected in the pattern of CDR being greatest in Chinese Americans, followed by Latinos, and African Americans having the smallest CDR. Further studies, especially population-based studies, are needed to confirm our findings.
Consistent with previous studies,7,34 we also found that longer AL was associated with thinner average and quadrant-specific RNFL in all racial/ethnic groups. When examined by ethnicity, AL was longest in Chinese Americans, followed by African Americans, and Latinos, with a mean of 23.7, 23.5, and 23.2 mm, respectively, similar to those reported by other studies.2 After accounting for the differences in AL, the difference in RNFL thickness between ethnicities was magnified.
We also found that reference limits, e.g. 5th percentile, for RNFL thickness in normal eyes differed by race/ethnicity. For example, the 5th percentile for RNFL thickness in the inferior quadrant in African Americans, Latino Americans, Chinese Americans, and all three groups combined was 91, 100, 99, and 97 μm, respectively. Using reference limits derived from a database composed of samples of different race/ethnicities would have resulted in more normal eyes in African American but less normal eyes in Chinese or Latino Americans being misclassified as abnormal. As the average RNFL thickness is thinner in European descendants than African descendants, by 0.34 μm in the Stratus OCT normative database7 and 3.8 μm in the Cirrus OCT Normative database,2 we expect the reference limits to be lower in European descendants. Because the normative database used in most OCT instruments in the U.S. are composed of primarily non-Hispanic whites,7 use of OCT-measured RNFL thickness in detecting early glaucoma among minority Americans needs to be interpreted with caution. Further studies are needed to investigate the diagnostic accuracy of RNFL reference databases in differentiating healthy and early glaucoma eyes among less-studied minority populations.
There are many strengths to note in this study. We utilized data from three similarly designed, population-based eye studies and obtained a very large sample size of OCT characterized eyes from understudied minority Americans. Because data for our analysis were collected from population-based studies, our conclusions may be more generalizable than those from clinical-based studies. However, the study also has a few limitations. First, with respect to the study population, we included only adults 50 years and older and our Latino population primarily consists of Mexican Americans. Direct comparison to non-Hispanic whites was not available because of the lack of similarly collected data from this population. While this study was unable to examine the differences between European and minority populations, we have described substantial differences in RNFL thickness among the three racial/ethnic groups in our study alone. We expect greater differences when comparing these groups to Europeans. In addition, our participants self-reported their race/ethnicity. While this is the most applicable approach, there may be concerns with validity. However, our findings are appropriate to generalize what encompasses both genetic and environmental components of race/ethnicity. Second, we did not include some factors that have been associated with glaucoma risk, such as central corneal thickness, in our multivariable analyses. However, CCT was not associated with RNFL in our study population. Third, our estimates of age-related loss in RNFL thickness were based on cross-sectional comparison of different age groups and therefore it may not reflect the actual longitudinal change that occurs. In fact, longitudinal data from the Advanced Imaging for Glaucoma Study (AIGS) study found that only 0.14 μm of the RNFL thickness was lost per year within a population comprising 90% of European descent.36 However, our estimates are comparable with the reported loss of 0.15-0.38μm in RNFL thickness per year from other cross-sectional reports.25,37,38, prospective studies that monitor longitudinal changes in RNFL thickness in population-based samples are still needed to confirm our observations.
In conclusion, we provided evidence on the relationship between demographic and clinical determinants of RNFL thickness and reported important differences in RNFL thickness among African, Chinese, and Latino Americans. Race/ethnicity was the largest contributor to the variability of average RNFL thickness and the race/ethnicity difference persists even after accounting for differences in AL and optic disc size, indicating that genetics may play a role in determining RNFL thickness. These findings highlight the need for consideration of race/ethnicity in clinical and research use of OCT-based RNFL measurement.
Supplementary Material
Retinal nerve fiber layer thickness differs across healthy African Americans, Chinese Americans, and Latinos racial/ethnic groups. Clinical applications of optical coherence tomography imaging should consider the role of race.
Acknowledgments
Financial Support: National Eye Institute, Grant/Award Numbers: U10EY011753, U10EY017337, U10EY023575, R21EY028721, National Institute of Environmental Health Sciences, Grant/Award Number: T32ES013678
The sponsor or funding organization had no role in the design or conduct of this research
Abbreviations
- RNFL
Retinal nerve fiber layer
- ONH
Optic nerve head
- OCT
Optical coherence tomography
- LALES
Los Angeles Latino Eye Study
- CHES
Chinese American Eye Study
- AFEDS
African Eye Disease Study
- AL
Axial length
- OD
Right
- OS
Left
- CDR
Cup-to-disc ratio
- VCDR
Vertical cup-to-disc ratio
- IOP
Intraocular pressure
- LMM
Linear mixed models
- SR2
Square semipartial correlation coefficient
- SEED
Singapore Epidemiology of Eye Diseases
- ADAGES
African Descent and Glaucoma Evaluation Study
- μm
Micrometer
- AIGS
Advanced Imaging for Glaucoma Study
Footnotes
Conflict of Interest: No conflicting relationship exists for any author
Meeting Presentation: The Association for Research in Vision and Ophthalmology Annual Meeting, 2017
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Alamouti B, Funk J. Retinal thickness decreases with age: An OCT study. Br J Ophthalmol. 2003;87(7):899–901. doi: 10.1136/bjo.87.7.899 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Knight OJ, Girkin CA, Budenz DL, Durbin MK, Feuer WJ. Effect of race, age, and axial length on optic nerve head parameters and retinal nerve fiber layer thickness measured by cirrus HD-OCT. Arch Ophthalmol. 2012;130(3):312–318. doi: 10.1001/archopthalmol.2011.1576 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Varma R, Skaf M, Barron E. Retinal nerve fiber layer thickness in normal human eyes. Ophthalmology. 1996;103(12):2114–2119. doi: 10.1016/S0161-6420(96)30381-3 [DOI] [PubMed] [Google Scholar]
- 4.Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science. 1991;254(5035):1178–1181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Health C for D and R. Workshops & Conferences (Medical Devices) - Public Workshop - FDA/American Glaucoma Society Workshop on the Validity, Reliability, and Usability of Glaucoma Imaging Devices, October 5, 2012. Center for Devices and Radiological Health; https://www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm318305.htm#date. Accessed October 18, 2018. [Google Scholar]
- 6.Realini T, Zangwill LM, Flanagan JG, et al. Normative Databases for Imaging Instrumentation. J Glaucoma. 2015;24(6):480–483. doi: 10.1097/IJG.0000000000000152 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Budenz DL, Anderson DR, Varma R, et al. Determinants of Normal Retinal Nerve Fiber Layer Thickness Measured by Stratus OCT. Ophthalmology. 2007;114(6):1046–1052. doi: 10.1016/j.ophtha.2006.08.046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Girkin CA, McGwin G, Sinai MJ, et al. Variation in optic nerve and macular structure with age and race with spectral-domain optical coherence tomography. Ophthalmology. 2011;118(12):2403–2408. doi: 10.1016/j.ophtha.2011.06.013 [DOI] [PubMed] [Google Scholar]
- 9.Chi T, Ritch R, Tsai C, Stickler D, Pitman B, Hsieh FY. Racial differences in optic nerve head parameters. Arch Ophthalmol. 1989;107(6):836–839. doi: 10.1001/archopht.1989.01070010858029 [DOI] [PubMed] [Google Scholar]
- 10.Racette L, Boden C, Kleinhandler SL, et al. Differences in visual function and optic nerve structure between healthy eyes of blacks and whites. Arch Ophthalmol. 2005;123(11):1547–1553. doi: 10.1001/archopht.123.11.1547 [DOI] [PubMed] [Google Scholar]
- 11.Marsh BC, Cantor LB, Wudunn D, et al. Optic nerve head (ONH) topographic analysis by stratus OCT in normal subjects: Correlation to disc size, age, and ethnicity. J Glaucoma. 2010;19(5):310–318. doi: 10.1097/IJG.0b013e3181b6e5cd [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Seider MI, Lee RY, Wang D, Pekmezci M, Porco TC, Lin SC. Optic disk size variability between African, Asian, white, hispanic, and filipino Americans using heidelberg retinal tomography. J Glaucoma. 2009;18(8):595–600. doi: 10.1097/IJG.0b013e3181996f05 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Biswas S, Lin C, Leung CKS. Evaluation of a myopic normative database for analysis of retinal nerve fiber layer thickness. JAMA Ophthalmol. 2016;134(9):1032–1039. doi: 10.1001/jamaophthalmol.2016.2343 [DOI] [PubMed] [Google Scholar]
- 14.Mwanza J-C, Oakley JD, Budenz DL, Anderson DR, Group TCOCTNDS. Ability of Cirrus™ HD-OCT Optic Nerve Head Parameters to Discriminate Normal from Glaucomatous Eyes. Ophthalmology. 2011;118(2):241–248.e1. doi: 10.1016/j.ophtha.2010.06.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Varma R, Paz SH, Azen SP, et al. The Los Angeles Latino Eye Study: design, methods, and baseline data. Ophthalmology. 2004;111(6):1121–1131. doi: 10.1016/j.ophtha.2004.02.001 [DOI] [PubMed] [Google Scholar]
- 16.Varma R, Hsu C, Wang D, Torres M, Azen SP. The Chinese American eye study: Design and methods. Ophthalmic Epidemiol. 2013;20(6):335–347. doi: 10.3109/09286586.2013.823505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McKean-Cowdin R, Fairbrother-Crisp A, Torres M, et al. The African American Eye Disease Study: Design and Methods. Ophthalmic Epidemiol. 2018;25(4):306–314. doi: 10.1080/09286586.2018.1454965 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Garcia-Martin E, Pinilla I, Idoipe M. Intra and interoperator reproducibility of retinal nerve fibre and macular thickness … Acta. 2010:3768. doi: 10.1111/j.1755-3768.2010.02045.x [DOI] [PubMed] [Google Scholar]
- 19.Keane PA, Mand PS, Liakopoulos S, Walsh AC, Sadda SR. Accuracy of retinal thickness measurements obtained with Cirrus optical coherence tomography. Br J Ophthalmol. 2009;93(11):1461–1467. doi: 10.1136/bjo.2008.155846 [DOI] [PubMed] [Google Scholar]
- 20.Mwanza JC, Durbin MK, Budenz DL. Interocular symmetry in peripapillary retinal nerve fiber layer thickness measured with the cirrus HD-OCT in healthy eyes. Am J Ophthalmol. 2011;151(3). doi: 10.1016/j.ajo.2010.09.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Budenz DL. Symmetry between the right and left eyes of the normal retinal nerve fiber layer measured with optical coherence tomography (an AOS thesis). Trans Am Ophthalmol Soc. 2008;106:252–275. /pmc/articles/PMC2646446/?report=abstract. Accessed September 12, 2020. [PMC free article] [PubMed] [Google Scholar]
- 22.Cheung CY, Chen D, Wong TY, et al. Determinants of quantitative optic nerve measurements using spectral domain optical coherence tomography in a population-based sample of non-glaucomatous subjects. Investig Ophthalmol Vis Sci. 2011;52(13):9629–9635. doi: 10.1167/iovs.11-7481 [DOI] [PubMed] [Google Scholar]
- 23.Khawaja AP, Chan MPY, Garway-Heath DF, et al. Associations with retinal nerve fiber layer measures in the EPIC-Norfolk Eye Study. Invest Ophthalmol Vis Sci. 2013;54(7):5028–5034. doi: 10.1167/iovs.13-11971 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ho H, Tham YC, Chee ML, et al. Retinal Nerve Fiber Layer Thickness in a Multiethnic Normal Asian Population: The Singapore Epidemiology of Eye Diseases Study. Ophthalmology. 2019;126(5):702–711. doi: 10.1016/j.ophtha.2018.11.031 [DOI] [PubMed] [Google Scholar]
- 25.Mauschitz MM, Bonnemaijer PWM, Diers K, et al. Systemic and Ocular Determinants of Peripapillary Retinal Nerve Fiber Layer Thickness Measurements in the European Eye Epidemiology (E3) Population. Ophthalmology. 2018;125(10):1526–1536. doi: 10.1016/j.ophtha.2018.03.026 [DOI] [PubMed] [Google Scholar]
- 26.Tao Y, Tham YC, Chee ML, et al. Profile of retinal nerve fibre layer symmetry in a multiethnic Asian population: The Singapore Epidemiology of Eye Diseases study. Br J Ophthalmol. 2019. doi: 10.1136/bjophthalmol-2019-314894 [DOI] [PubMed] [Google Scholar]
- 27.Kashani AH, Zimmer-Galler IE, Shah SM, et al. Retinal Thickness Analysis by Race, Gender, and Age Using Stratus OCT. Am J Ophthalmol. 2010;149(3):496–502.e1. doi: 10.1016/j.ajo.2009.09.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Varma R, Ying-Lai M, Francis BA, et al. Prevalence of open-angle glaucoma and ocular hypertension in Latinos: The Los Angeles Latino Eye Study. Ophthalmology. 2004;111(8):1439–1448. doi: 10.1016/j.ophtha.2004.01.025 [DOI] [PubMed] [Google Scholar]
- 29.Leske MC. Open-angle glaucoma - An epidemiologic overview. In: Ophthalmic Epidemiology. Vol 14. ; 2007:166–172. doi: 10.1080/09286580701501931 [DOI] [PubMed] [Google Scholar]
- 30.Tielsch JM, Sommer A, Katz J, Royall RM, Quigley HA, Javitt J. Racial Variations in the Prevalence of Primary Open-angle Glaucoma: The Baltimore Eye Survey. JAMA J Am Med Assoc. 1991;266(3):369–374. doi: 10.1001/jama.1991.03470030069026 [DOI] [PubMed] [Google Scholar]
- 31.Varma R, Wang D, Wu C, et al. Four-year incidence of open-angle glaucoma and ocular hypertension: The Los Angeles Latino Eye Study. Am J Ophthalmol. 2012;154(2). doi: 10.1016/j.ajo.2012.02.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Cho H kyung, Kee C. Population-based glaucoma prevalence studies in Asians. Surv Ophthalmol. 2014;59(4):434–447. doi: 10.1016/j.survophthal.2013.09.003 [DOI] [PubMed] [Google Scholar]
- 33.Funaki S, Shirakashi M, Abe H. Relation between size of optic disc and thickness of retinal nerve fibre layer in normal subjects. Br J Ophthalmol. 1998;82(11):1242–1245. doi: 10.1136/bjo.82.11.1242 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Oliveira C, Harizman N, Girkin CA, et al. Axial length and optic disc size in normal eyes. Br J Ophthalmol. 2007;91(1):37–39. doi: 10.1136/bjo.2006.102061 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Varma R, Tielsch JM, Quigley HA, et al. Race-, Age-, Gender-, and Refractive Error—Related Differences in the Normal Optic Disc. Arch Ophthalmol. 1994;112(8):1068–1076. doi: 10.1001/archopht.1994.01090200074026 [DOI] [PubMed] [Google Scholar]
- 36.Zhang X, Francis BA, Dastiridou A, et al. Longitudinal and Cross-Sectional Analyses of Age Effects on Retinal Nerve Fiber Layer and Ganglion Cell Complex Thickness by Fourier-Domain OCT for the Advanced Imaging for Glaucoma Study Group. Trans Vis Sci Tech. 2016;5(2):1–9. doi: 10.1167/tvst.5.2.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Patel NB, Lim M, Gajjar A, Evans KB, Harwerth RS. Age-associated changes in the retinal nerve fiber layer and optic nerve head. Investig Ophthalmol Vis Sci. 2014;55(8):5134–5143. doi: 10.1167/iovs.14-14303 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bowd C, Zangwill LM, Blumenthal EZ, et al. Imaging of the optic disc and retinal nerve fiber layer: the effects of age, optic disc area, refractive error, and gender. J Opt Soc Am Opt image Sci vision. 2002;19(1):197–207. doi: 10.1364/JOSAA.19.000197 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.